Andrew, J, Kaidonis, MA & Stoianoff, NP 2016, 'The Shifting Meaning of Sustainability' in Aras, G & Crowther, D (eds), A Handbook of Corporate Governance and Social Responsibility, Gower Publishing Limited, England, pp. 83-90.
View/Download from: Publisher's site
View description>>
Sustainability, as a concept, gained momentum as international non-government organizations developed the term. The United Nations’ Brundtland Report is credited with first referring to sustainability as having three necessary and coexisting components being, environmental, economic and social sustainability. International accounting professional institutions also responded to this momentum, at first with an in principle adoption of the term. As sustainability reporting accompanied financial reporting, the concepts of business were also imposed on the term. The objective of global equity was surpassed by financial terminology which also prioritized concepts of risks and opportunities to explore market potentials.
de la Cruz del Río-Rama, M, Peris-Ortiz, M & Merigó-Lindahl, JM 2016, 'Monterrei Wine Tourist Route (Galicia-Spain): Analysis from the Perspective of Offer' in Wine and Tourism, Springer International Publishing, pp. 57-69.
View/Download from: Publisher's site
View description>>
© Springer International Publishing Switzerland 2016. In this chapter, the results of an empirical study carried out in establishments (wineries, restaurants, and hotel accommodation) adhered to the Monterrei Wine Route (Galicia-Spain) in 2013 are presented. The objectives of the study were to know the profi le of establishments and the offer of enotourism activities, as well as to analyze the profi le of enotourists from the point of view of the establishments, to measure the level of satisfaction of their adherence to the route, the perceived image of the destination, and the assessment of the enotourism activity of the route. Finally, a SWOT analysis is carried out from the information provided by the establishments that took part in the survey. The methodology consists on a descriptive analysis. The research results have let us respond to all the proposed objectives, showing the adhered establishments of a low level of satisfaction with their adherence to the route.
Guglyuvatyy, E & Stoianoff, NP 2016, 'Carbon Policy in Australia – A Political History' in Stoianoff, NP, Kreiser, L, Butcher, B, Milne, JE & Ashiabor, H (eds), Green Fiscal Reform for a Sustainable Future Reform, Innovation and Renewable Energy, Edward Elgar Publishing, Cheltenham UK, pp. 31-52.
View/Download from: Publisher's site
View description>>
Australia had actively participated in the 1992 Earth Summit in Rio de Janeiro, endorsing the Summit goals which were formed by the desire for sustainable development. Australia also joined the United Nations Framework Convention on Climate Change and much later signed the Kyoto Protocol enthusiastically supporting greenhouse gas reduction. A range of measures aimed to reduce Australia’s greenhouse gas emissions have been on the agenda at the Federal and State level for the last two decades. Until recently, successive Australian governments have been committed to the introduction of a carbon tax or an emissions trading scheme designed to mitigate climate change. This paper examines the historical progress of Australian climate change policy including the implementation of the present Australian Government’s Direct Action Plan. The article in particular observes several interesting and significant aspects of Australian climate law highlighting governmental approaches and processes leading to the introduction of those laws. The historical perspective is necessary to identify most common features of the climate law implementation procedures and to identify what political factors influence these processes in Australia. Examination of the Australian climate change regime indicates how different actors influence policy proposals to achieve their own goals, rather than to cooperate in a process of generating the best overall legal option. This paper concludes that the development of climate law in Australia required some innovative and responsive law initiatives. However, the practical implementation of various climate change laws had been constantly impacted by various economic and political factors.
Guglyuvatyy, E & Stoianoff, NP 2016, 'Carbon policy in Australia – a political history' in Green Fiscal Reform for a Sustainable Future, Edward Elgar Publishing.
View/Download from: Publisher's site
Stoianoff, NP 2016, 'Ensuring a Sustainable Future Through Recognizing and Protecting Indigenous Ecological Knowledge' in Mauerhofer, V (ed), Legal Aspects of Sustainable Development, Springer International Publishing, Switzerland, pp. 109-123.
View/Download from: Publisher's site
View description>>
This paper sets out the way in which Indigenous ecological knowledge has received increasing recognition as a holistic mechanism through which Australia’s natural resources can be sustainably managed. This increased recognition and consequent utilization needs to take place within a legal framework that acknowledges and respects the customary laws and rules of the Indigenous ecological knowledge holders and provides appropriate benefits back to those knowledge holders. This paper considers the nature of such a legal framework and reports on the research conducted by the author and her research team, through the use of action research and Indigenous research paradigm methodologies, in developing such a legal regime that encapsulates the principles established in the Convention on Biological Diversity 1992, expanded in the Nagoya Protocol to the Convention, and reinforced in the United Nations Declaration on the Rights of Indigenous Peoples 2007. The result was a White Paper espousing a sui generis legal framework of recognition and protection of Indigenous knowledge associated with natural resource management focussed on the Aboriginal Communities of the state of New South Wales in Australia and accordingly reflects the concerns and interests of those communities while incorporating the international law principles described above. This was achieved through an initial comparative analysis of regimes already in existence in other nations, the establishment of a highly skilled and multidisciplinary Working Party representing both Indigenous and non-Indigenous individuals and stakeholders, and finally through Aboriginal Community consultation.
Stoianoff, NP & Kumari, TV 2016, 'Intellectual Property' in Star, S (ed), Australia and India - A Comparative Overview of the Law and Legal Practice, Universal Law Publishing Co. Ltd., India, pp. 235-264.
Zhang, Y & Xu, G 2016, 'Singular Value Decomposition' in Encyclopedia of Database Systems, Springer New York, pp. 1-3.
View/Download from: Publisher's site
Ahadi, A, Brennan, S, Kennedy, PJ, Hutvagner, G & Tran, N 2016, 'Long non-coding RNAs harboring miRNA seed regions are enriched in prostate cancer exosomes', Scientific Reports, vol. 6, no. 1, pp. 1-14.
View/Download from: Publisher's site
View description>>
AbstractLong non-coding RNAs (lncRNAs) form the largest transcript class in the human transcriptome. These lncRNA are expressed not only in the cells, but they are also present in the cell-derived extracellular vesicles such as exosomes. The function of these lncRNAs in cancer biology is not entirely clear, but they appear to be modulators of gene expression. In this study, we characterize the expression of lncRNAs in several prostate cancer exosomes and their parental cell lines. We show that certain lncRNAs are enriched in cancer exosomes with the overall expression signatures varying across cell lines. These exosomal lncRNAs are themselves enriched for miRNA seeds with a preference for let-7 family members as well as miR-17, miR-18a, miR-20a, miR-93 and miR-106b. The enrichment of miRNA seed regions in exosomal lncRNAs is matched with a concomitant high expression of the same miRNA. In addition, the exosomal lncRNAs also showed an over representation of RNA binding protein binding motifs. The two most common motifs belonged to ELAVL1 and RBMX. Given the enrichment of miRNA and RBP sites on exosomal lncRNAs, their interplay may suggest a possible function in prostate cancer carcinogenesis.
Al-Jubouri, B & Gabrys, B 2016, 'Local Learning for Multi-layer, Multi-component Predictive System', Procedia Computer Science, vol. 96, pp. 723-732.
View/Download from: Publisher's site
View description>>
This study introduces a new multi-layer multi-component ensemble. The components of this ensemble are trained locally on subsets of features for disjoint sets of data. The data instances are assigned to local regions using the similarity of their features pairwise squared correlation. Many ensemble methods encourage diversity among their base predictors by training them on different subsets of data or different subsets of features. In the proposed architecture the local regions contain disjoint sets of data and for this data only the most similar features are selected. The pairwise squared correlations of the features are used to weight the predictions of the ensemble's models. The proposed architecture has been tested on a number of data sets and its performance was compared to five benchmark algorithms. The results showed that the testing accuracy of the developed architecture is comparable to the rotation forest and is better than the other benchmark algorithms.
Anaissi, A, Goyal, M, Catchpoole, DR, Braytee, A & Kennedy, PJ 2016, 'Ensemble Feature Learning of Genomic Data Using Support Vector Machine', PLOS ONE, vol. 11, no. 6, pp. e0157330-e0157330.
View/Download from: Publisher's site
View description>>
© 2016 Anaissi et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. The identification of a subset of genes having the ability to capture the necessary information to distinguish classes of patients is crucial in bioinformatics applications. Ensemble and bagging methods have been shown to work effectively in the process of gene selection and classification. Testament to that is random forest which combines random decision trees with bagging to improve overall feature selection and classification accuracy. Surprisingly, the adoption of these methods in support vector machines has only recently received attention but mostly on classification not gene selection. This paper introduces an ensemble SVM-Recursive Feature Elimination (ESVM-RFE) for gene selection that follows the concepts of ensemble and bagging used in random forest but adopts the backward elimination strategy which is the rationale of RFE algorithm. The rationale behind this is, building ensemble SVM models using randomly drawn bootstrap samples from the training set, will produce different feature rankings which will be subsequently aggregated as one feature ranking. As a result, the decision for elimination of features is based upon the ranking of multiple SVM models instead of choosing one particular model. Moreover, this approach will address the problem of imbalanced datasets by constructing a nearly balanced bootstrap sample. Our experiments show that ESVM-RFE for gene selection substantially increased the classification performance on five microarray datasets compared to state-of-the-art methods. Experiments on the childhood leukaemia dataset show that an average 9% better accuracy is achieved by ESVM-RFE over SVM-RFE, and 5% over random forest based approach. The selected genes by the ESVM-RFE algo...
Avellino, R, Havermans, M, Erpelinck, C, Sanders, MA, Hoogenboezem, R, van de Werken, HJG, Rombouts, E, van Lom, K, van Strien, PMH, Gebhard, C, Rehli, M, Pimanda, J, Beck, D, Erkeland, S, Kuiken, T, de Looper, H, Gröschel, S, Touw, I, Bindels, E & Delwel, R 2016, 'An autonomous CEBPA enhancer specific for myeloid-lineage priming and neutrophilic differentiation', Blood, vol. 127, no. 24, pp. 2991-3003.
View/Download from: Publisher's site
View description>>
Key Points The CEBPA locus harbors 14 enhancers of which distinct combinations are active in different CEBPA-expressing tissues. A +42-kb enhancer is required for myeloid-lineage priming to drive adequate CEBPA expression levels necessary for neutrophilic maturation.
Benavides Espinosa, MDM & Merigó Lindahl, JM 2016, 'Organizational design as a learning enabler: A fuzzy-set approach', Journal of Business Research, vol. 69, no. 4, pp. 1340-1344.
View/Download from: Publisher's site
View description>>
In the literature on organizational learning, very few empirical studies attempt to show how organizational design can enable or hinder learning in organizations. This study uses a fuzzy-set technique (fuzzy-set qualitative comparative analysis: fsQCA) as an initial approach to analyzing different design variables and how they affect organizational learning. The results prove that themechanical structures are suitable for organizational learning, especially in large companies. Furthermore, qualified workers should have autonomy to learn.
Blanco-Mesa, F, Merigó, JM & Kacprzyk, J 2016, 'Bonferroni means with distance measures and the adequacy coefficient in entrepreneurial group theory', Knowledge-Based Systems, vol. 111, pp. 217-227.
View/Download from: Publisher's site
View description>>
© 2016 The aim of the paper is to develop new aggregation operators using Bonferroni means, OWA operators and some distance measure. We introduce the BON-OWAAC and BON-OWAIMAM operators. We are able to include coefficient adequacy and the maximum and minimum levels in the same formulation with Bonferroni means and an OWA operator. The main advantages of using these operators are that they allow consideration of continuous aggregations, multiple comparisons between each argument and distance measures in the same formulation. An application is developed using these new algorithms in combination with Moore's families and Galois lattices to solve group decision-making problems. The professional and personal interests of the entrepreneurs who share co-working spaces are taken as an example for establishing relationships and groups. According to the professional and personal profile affinities for each entrepreneur, the results show dissimilarity and fuzzy relationships and the maximum similarity sub-relations to establish relationships and groups using Moore's families and Galois lattice. Finally, this new type of distance family can be used for applications in areas such as sports teams, strategy marketing and teamwork.
Botezatu, L, Michel, LC, Makishima, H, Schroeder, T, Germing, U, Haas, R, van der Reijden, B, Marneth, AE, Bergevoet, SM, Jansen, JH, Przychodzen, B, Wlodarski, M, Niemeyer, C, Platzbecker, U, Ehninger, G, Unnikrishnan, A, Beck, D, Pimanda, J, Hellström-Lindberg, E, Malcovati, L, Boultwood, J, Pellagatti, A, Papaemmanuil, E, Le Coutre, P, Kaeda, J, Opalka, B, Möröy, T, Dührsen, U, Maciejewski, J & Khandanpour, C 2016, 'GFI136N as a therapeutic and prognostic marker for myelodysplastic syndrome', Experimental Hematology, vol. 44, no. 7, pp. 590-595.e1.
View/Download from: Publisher's site
Cao, L 2016, 'Data science and analytics: a new era', International Journal of Data Science and Analytics, vol. 1, no. 1, pp. 1-2.
View/Download from: Publisher's site
Cao, L, Dong, X & Zheng, Z 2016, 'e-NSP: Efficient negative sequential pattern mining', Artificial Intelligence, vol. 235, pp. 156-182.
View/Download from: Publisher's site
View description>>
© 2016 The Authors. Published by Elsevier B.V. As an important tool for behavior informatics, negative sequential patterns (NSP) (such as missing medical treatments) are critical and sometimes much more informative than positive sequential patterns (PSP) (e.g. using a medical service) in many intelligent systems and applications such as intelligent transport systems, healthcare and risk management, as they often involve non-occurring but interesting behaviors. However, discovering NSP is much more difficult than identifying PSP due to the significant problem complexity caused by non-occurring elements, high computational cost and huge search space in calculating negative sequential candidates (NSC). So far, the problem has not been formalized well, and very few approaches have been proposed to mine for specific types of NSP, which rely on database re-scans after identifying PSP in order to calculate the NSC supports. This has been shown to be very inefficient or even impractical, since the NSC search space is usually huge. This paper proposes a very innovative and efficient theoretical framework: Set theory-based NSP mining (ST-NSP), and a corresponding algorithm, e-NSP, to efficiently identify NSP by involving only the identified PSP, without re-scanning the database. Accordingly, negative containment is first defined to determine whether a data sequence contains a negative sequence based on set theory. Second, an efficient approach is proposed to convert the negative containment problem to a positive containment problem. The NSC supports are then calculated based only on the corresponding PSP. This not only avoids the need for additional database scans, but also enables the use of existing PSP mining algorithms to mine for NSP. Finally, a simple but efficient strategy is proposed to generate NSC. Theoretical analyses show that e-NSP performs particularly well on datasets with a small number of elements in a sequence, a large number of itemsets and low...
Casanovas, M, Torres-Martínez, A & Merigó, JM 2016, 'Decision Making in Reinsurance with Induced OWA Operators and Minkowski Distances', Cybernetics and Systems, vol. 47, no. 6, pp. 460-477.
View/Download from: Publisher's site
Chandrakanthan, V, Yeola, A, Kwan, JC, Oliver, RA, Qiao, Q, Kang, YC, Zarzour, P, Beck, D, Boelen, L, Unnikrishnan, A, Villanueva, JE, Nunez, AC, Knezevic, K, Palu, C, Nasrallah, R, Carnell, M, Macmillan, A, Whan, R, Yu, Y, Hardy, P, Grey, ST, Gladbach, A, Delerue, F, Ittner, L, Mobbs, R, Walkley, CR, Purton, LE, Ward, RL, Wong, JWH, Hesson, LB, Walsh, W & Pimanda, JE 2016, 'PDGF-AB and 5-Azacytidine induce conversion of somatic cells into tissue-regenerative multipotent stem cells', PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA, vol. 113, no. 16, pp. E2306-E2315.
View/Download from: Publisher's site
Chen, Q, Chen, Y-PP & Zhang, C 2016, 'Interval-Based Similarity for Classifying Conserved RNA Secondary Structures', IEEE Intelligent Systems, vol. 31, no. 3, pp. 78-85.
View/Download from: Publisher's site
Chua, T-S, He, X, Liu, W, Piccardi, M, Wen, Y & Tao, D 2016, 'Big data meets multimedia analytics.', Signal Process., vol. 124, pp. 1-4.
View/Download from: Publisher's site
Cramb, SM, Mengersen, KL, Lambert, PC, Ryan, LM & Baade, PD 2016, 'A flexible parametric approach to examining spatial variation in relative survival', Statistics in Medicine, vol. 35, no. 29, pp. 5448-5463.
View/Download from: Publisher's site
View description>>
Most of the few published models used to obtain small‐area estimates of relative survival are based on a generalized linear model with piecewise constant hazards under a Bayesian formulation. Limitations of these models include the need to artificially split the time scale, restricted ability to include continuous covariates, and limited predictive capacity. Here, an alternative Bayesian approach is proposed: a spatial flexible parametric relative survival model. This overcomes previous limitations by combining the benefits of flexible parametric models: the smooth, well‐fitting baseline hazard functions and predictive ability, with the Bayesian benefits of robust and reliable small‐area estimates. Both spatially structured and unstructured frailty components are included. Spatial smoothing is conducted using the intrinsic conditional autoregressive prior. The model was applied to breast, colorectal, and lung cancer data from the Queensland Cancer Registry across 478 geographical areas. Advantages of this approach include the ease of including more realistic complexity, the feasibility of using individual‐level input data, and the capacity to conduct overall, cause‐specific, and relative survival analysis within the same framework. Spatial flexible parametric survival models have great potential for exploring small‐area survival inequalities, and we hope to stimulate further use of these models within wider contexts. Copyright © 2016 John Wiley & Sons, Ltd.
Cui, P, Liu, H, Aggarwal, C & Wang, F 2016, 'Uncovering and Predicting Human Behaviors', IEEE Intelligent Systems, vol. 31, no. 2, pp. 77-88.
View/Download from: Publisher's site
Ghosh, S, Feng, M, Nguyen, H & Li, J 2016, 'Hypotension Risk Prediction via Sequential Contrast Patterns of ICU Blood Pressure', IEEE Journal of Biomedical and Health Informatics, vol. 20, no. 5, pp. 1416-1426.
View/Download from: Publisher's site
View description>>
© 2013 IEEE. Acute hypotension is a significant risk factor for in-hospital mortality at intensive care units. Prolonged hypotension can cause tissue hypoperfusion, leading to cellular dysfunction and severe injuries to multiple organs. Prompt medical interventions are thus extremely important for dealing with acute hypotensive episodes (AHE). Population level prognostic scoring systems for risk stratification of patients are suboptimal in such scenarios. However, the design of an efficient risk prediction system can significantly help in the identification of critical care patients, who are at risk of developing an AHE within a future time span. Toward this objective, a pattern mining algorithm is employed to extract informative sequential contrast patterns from hemodynamic data, for the prediction of hypotensive episodes. The hypotensive and normotensive patient groups are extracted from the MIMIC-II critical care research database, following an appropriate clinical inclusion criteria. The proposed method consists of a data preprocessing step to convert the blood pressure time series into symbolic sequences, using a symbolic aggregate approximation algorithm. Then, distinguishing subsequences are identified using the sequential contrast mining algorithm. These subsequences are used to predict the occurrence of an AHE in a future time window separated by a user-defined gap interval. Results indicate that the method performs well in terms of the prediction performance as well as in the generation of sequential patterns of clinical significance. Hence, the novelty of sequential patterns is in their usefulness as potential physiological biomarkers for building optimal patient risk stratification systems and for further clinical investigation of interesting patterns in critical care patients.
Gong, C, Tao, D, Maybank, SJ, Liu, W, Kang, G & Yang, J 2016, 'Multi-Modal Curriculum Learning for Semi-Supervised Image Classification', IEEE TRANSACTIONS ON IMAGE PROCESSING, vol. 25, no. 7, pp. 3249-3260.
View/Download from: Publisher's site
View description>>
© 1992-2012 IEEE. Semi-supervised image classification aims to classify a large quantity of unlabeled images by typically harnessing scarce labeled images. Existing semi-supervised methods often suffer from inadequate classification accuracy when encountering difficult yet critical images, such as outliers, because they treat all unlabeled images equally and conduct classifications in an imperfectly ordered sequence. In this paper, we employ the curriculum learning methodology by investigating the difficulty of classifying every unlabeled image. The reliability and the discriminability of these unlabeled images are particularly investigated for evaluating their difficulty. As a result, an optimized image sequence is generated during the iterative propagations, and the unlabeled images are logically classified from simple to difficult. Furthermore, since images are usually characterized by multiple visual feature descriptors, we associate each kind of features with a teacher, and design a multi-modal curriculum learning (MMCL) strategy to integrate the information from different feature modalities. In each propagation, each teacher analyzes the difficulties of the currently unlabeled images from its own modality viewpoint. A consensus is subsequently reached among all the teachers, determining the currently simplest images (i.e., a curriculum), which are to be reliably classified by the multi-modal learner. This well-organized propagation process leveraging multiple teachers and one learner enables our MMCL to outperform five state-of-the-art methods on eight popular image data sets.
González, LO, Rodríguez Gil, LI, Martorell Cunill, O & Merigó Lindahl, JM 2016, 'The effect of financial innovation on European banks' risk', Journal of Business Research, vol. 69, no. 11, pp. 4781-4786.
View/Download from: Publisher's site
View description>>
© 2016 Elsevier Inc. This study examines the effect of the use of securitization and credit derivatives on the risk profile of European banks. Using information from 134 listed European banks during the period of 2006–2010, the results show that securitization and trading with credit derivatives have a negative effect on financial stability. The main findings also show the dominance of trading positions over hedging positions for credit derivatives. The results of this study support the higher capital requirements of the new Basel III international banking regulations. Furthermore, accounting measures do not readily indicate market risks, and thus the results support central banks’ use of market-solvency measures to monitor financial stability.
Hanh, LTM, Binh, NT & Tung, KT 2016, 'A Novel Fitness function of metaheuristic algorithms for test data generation for simulink models based on mutation analysis', Journal of Systems and Software, vol. 120, pp. 17-30.
View/Download from: Publisher's site
Hazber, MAG, Li, R, Gu, X & Xu, G 2016, 'Integration Mapping Rules: Transforming Relational Database to Semantic Web Ontology', Applied Mathematics & Information Sciences, vol. 10, no. 3, pp. 881-901.
View/Download from: Publisher's site
View description>>
© 2016 NSP. Semantic integration became an attractive area of research in several disciplines, such as information integration, databases and ontologies. Huge amount of data is still stored in relational databases (RDBs) that can be used to build ontology, and the database cannot be used directly by the semantic web. Therefore, one of the main challenges of the semantic web is mapping relational databases to ontologies (RDF(S)-OWL). Moreover, the use of manual work in the mapping of web contents to ontologies is impractical because it contains billions of pages and the most of these contents are generated from relational databases. Hence, we propose a new approach, which enables semantic web applications to access relational databases and their contents by semantic methods. Domain ontologies can be used to formulate relational database schema and data in order to simplify the mapping (transformation) of the underlying data sources. Our method consists of two main phases: building ontology from an RDB schema and the generation of ontology instances from an RDB data automatically. In the first phase, we studied different cases of RDB schema to be mapped into ontology represented in RDF(S)-OWL, while in the second phase, the mapping rules are used to transform RDB data to ontological instances represented in RDF triples. Our approach is demonstrated with examples, validated by ontology validator and implemented using Apache Jena in Java Language and MYSQL. This approach is effective for building ontology and important for mining semantic information from huge web resources.
He, W & Xu, G 2016, 'Social media analytics: unveiling the value, impact and implications of social media analytics for the management and use of online information', Online Information Review, vol. 40, no. 1.
View/Download from: Publisher's site
Hesson, LB, Ng, B, Zarzour, P, Srivastava, S, Kwok, C-T, Packham, D, Nunez, AC, Beck, D, Ryan, R, Dower, A, Ford, CE, Pimanda, JE, Sloane, MA, Hawkins, NJ, Bourke, MJ, Wong, JWH & Ward, RL 2016, 'Integrated Genetic, Epigenetic, and Transcriptional Profiling Identifies Molecular Pathways in the Development of Laterally Spreading Tumors', MOLECULAR CANCER RESEARCH, vol. 14, no. 12, pp. 1217-1228.
View/Download from: Publisher's site
Hou, S, Zhou, S, Chen, L, Feng, Y & Awudu, K 2016, 'Multi-label learning with label relevance in advertising video', Neurocomputing, vol. 171, pp. 932-948.
View/Download from: Publisher's site
View description>>
The recent proliferation of videos has brought out the need for applications such as automatic annotation and organization. These applications could greatly benefit from the respective thematic content depending on the type of video. Unlike the other kinds of video, an advertising video usually conveys a specific theme in a certain time period (e.g. drawing the audience׳s attention to a product or emphasizing the brand). Traditional multi-label algorithms may not work effectively with advertising videos due mainly to their heterogeneous nature. In this paper, we propose a new learning paradigm to resolve the problems arising out of traditional multi-label learning in advertising videos through label relevance. Aiming to address the issue of label relevance, we firstly assign each label with label degree (LD) to classify all the labels into three groups such as first label (FL), important label (IL) and common label (CL), and then propose a Directed Probability Label Graph (DPLG) model to mine the most related labels from the multi-label data with label relevance, in which the interdependency between labels is considered. In the implementation of DPLG, the labels that appear occasionally and possess inconspicuous co-occurrences are consequently eliminated effectively, employing λ-filtering and τ-pruning processes, respectively. And then the graph theory is utilized in DPLG to acquire Correlative Label-Sets (CLSs). Lastly, the searched Correlative Label-Sets (CLSs) are utilized to enhance multi-label annotation. Experimental results on advertising videos and several publicly available datasets demonstrate the effectiveness of the proposed method for multi-label annotation with label relevance
Huang, S, Zhang, J, Wang, L & Hua, X-S 2016, 'Social Friend Recommendation Based on Multiple Network Correlation', IEEE TRANSACTIONS ON MULTIMEDIA, vol. 18, no. 2, pp. 287-299.
View/Download from: Publisher's site
View description>>
© 2015 IEEE. Friend recommendation is an important recommender application in social media. Major social websites such as Twitter and Facebook are all capable of recommending friends to individuals. However, most of these websites use simple friend recommendation algorithms such as similarity, popularity, or 'friend's friends are friends,' which are intuitive but consider few of the characteristics of the social network. In this paper we investigate the structure of social networks and develop an algorithm for network correlation-based social friend recommendation (NC-based SFR). To accomplish this goal, we correlate different 'social role' networks, find their relationships and make friend recommendations. NC-based SFR is characterized by two key components: 1) related networks are aligned by selecting important features from each network, and 2) the network structure should be maximally preserved before and after network alignment. After important feature selection has been made, we recommend friends based on these features. We conduct experiments on the Flickr network, which contains more than ten thousand nodes and over 30 thousand tags covering half a million photos, to show that the proposed algorithm recommends friends more precisely than reference methods.
Huang, X, Zhang, J, Fan, L, Wu, Q & Yuan, C 2016, 'A Systematic Approach for Cross-source Point Cloud Registration by Preserving Macro and Micro Structures', IEEE Transactions on Image Processing, vol. 26, no. 7, pp. 3261-3276.
View/Download from: Publisher's site
View description>>
We propose a systematic approach for registering cross-source point clouds.The compelling need for cross-source point cloud registration is motivated bythe rapid development of a variety of 3D sensing techniques, but many existingregistration methods face critical challenges as a result of the largevariations in cross-source point clouds. This paper therefore illustrates anovel registration method which successfully aligns two cross-source pointclouds in the presence of significant missing data, large variations in pointdensity, scale difference and so on. The robustness of the method is attributedto the extraction of macro and micro structures. Our work has three maincontributions: (1) a systematic pipeline to deal with cross-source point cloudregistration; (2) a graph construction method to maintain macro and microstructures; (3) a new graph matching method is proposed which considers theglobal geometric constraint to robustly register these variable graphs.Compared to most of the related methods, the experiments show that the proposedmethod successfully registers in cross-source datasets, while other methodshave difficulty achieving satisfactory results. The proposed method also showsgreat ability in same-source datasets.
Huang, Y, Thoms, JAI, Tursky, ML, Knezevic, K, Beck, D, Chandrakanthan, V, Suryani, S, Olivier, J, Boulton, A, Glaros, EN, Thomas, SR, Lock, RB, MacKenzie, KL, Bushweller, JH, Wong, JWH & Pimanda, JE 2016, 'MAPK/ERK2 phosphorylates ERG at serine 283 in leukemic cells and promotes stem cell signatures and cell proliferation', LEUKEMIA, vol. 30, no. 7, pp. 1552-1561.
View/Download from: Publisher's site
View description>>
© 2016 Macmillan Publishers Limited. Aberrant ERG (v-ets avian erythroblastosis virus E26 oncogene homolog) expression drives leukemic transformation in mice and high expression is associated with poor patient outcomes in acute myeloid leukemia (AML) and T-acute lymphoblastic leukemia (T-ALL). Protein phosphorylation regulates the activity of many ETS factors but little is known about ERG in leukemic cells. To characterize ERG phosphorylation in leukemic cells, we applied liquid chromatography coupled tandem mass spectrometry and identified five phosphorylated serines on endogenous ERG in T-ALL and AML cells. S283 was distinct as it was abundantly phosphorylated in leukemic cells but not in healthy hematopoietic stem and progenitor cells (HSPCs). Overexpression of a phosphoactive mutant (S283D) increased expansion and clonogenicity of primary HSPCs over and above wild-type ERG. Using a custom antibody, we screened a panel of primary leukemic xenografts and showed that ERG S283 phosphorylation was mediated by mitogen-activated protein kinase/extracellular signal-regulated kinase (MAPK/ERK) signaling and in turn regulated expression of components of this pathway. S283 phosphorylation facilitates ERG enrichment and transactivation at the ERG +85 HSPC enhancer that is active in AML and T-ALL with poor prognosis. Taken together, we have identified a specific post-translational modification in leukemic cells that promotes progenitor proliferation and is a potential target to modulate ERG-driven transcriptional programs in leukemia.
Huque, MH, Anderson, C, Walton, R & Ryan, L 2016, 'Individual level covariate adjusted conditional autoregressive (indiCAR) model for disease mapping', International Journal of Health Geographics, vol. 15, no. 1, pp. 1-13.
View/Download from: Publisher's site
View description>>
© 2016 The Author(s). Background: Mapping disease rates over a region provides a visual illustration of underlying geographical variation of the disease and can be useful to generate new hypotheses on the disease aetiology. However, methods to fit the popular and widely used conditional autoregressive (CAR) models for disease mapping are not feasible in many applications due to memory constraints, particularly when the sample size is large. We propose a new algorithm to fit a CAR model that can accommodate both individual and group level covariates while adjusting for spatial correlation in the disease rates, termed indiCAR. Our method scales well and works in very large datasets where other methods fail. Results: We evaluate the performance of the indiCAR method through simulation studies. Our simulation results indicate that the indiCAR provides reliable estimates of all the regression and random effect parameters. We also apply indiCAR to the analysis of data on neutropenia admissions in New South Wales (NSW), Australia. Our analyses reveal that lower rates of neutropenia admissions are significantly associated with individual level predictors including higher age, male gender, residence in an outer regional area and a group level predictor of social disadvantage, the socio-economic index for areas. A large value for the spatial dependence parameter is estimated after adjusting for individual and area level covariates. This suggests the presence of important variation in the management of cancer patients across NSW. Conclusions: Incorporating individual covariate data in disease mapping studies improves the estimation of fixed and random effect parameters by utilizing information from multiple sources. Health registries routinely collect individual and area level information and thus could benefit by using indiCAR for mapping disease rates. Moreover, the natural applicability of indiCAR in a distributed computing framework enhances its application in...
Huque, MH, Bondell, HD, Carroll, RJ & Ryan, LM 2016, 'Spatial Regression with Covariate Measurement Error: A Semiparametric Approach', Biometrics, vol. 72, no. 3, pp. 678-686.
View/Download from: Publisher's site
View description>>
Summary Spatial data have become increasingly common in epidemiology and public health research thanks to advances in GIS (Geographic Information Systems) technology. In health research, for example, it is common for epidemiologists to incorporate geographically indexed data into their studies. In practice, however, the spatially defined covariates are often measured with error. Naive estimators of regression coefficients are attenuated if measurement error is ignored. Moreover, the classical measurement error theory is inapplicable in the context of spatial modeling because of the presence of spatial correlation among the observations. We propose a semiparametric regression approach to obtain bias-corrected estimates of regression parameters and derive their large sample properties. We evaluate the performance of the proposed method through simulation studies and illustrate using data on Ischemic Heart Disease (IHD). Both simulation and practical application demonstrate that the proposed method can be effective in practice.
Huque, MH, Carroll, RJ, Diao, N, Christiani, DC & Ryan, LM 2016, 'Exposure Enriched Case‐Control (EECC) Design for the Assessment of Gene–Environment Interaction', Genetic Epidemiology, vol. 40, no. 7, pp. 570-578.
View/Download from: Publisher's site
View description>>
ABSTRACTGenetic susceptibility and environmental exposure both play an important role in the aetiology of many diseases. Case‐control studies are often the first choice to explore the joint influence of genetic and environmental factors on the risk of developing a rare disease. In practice, however, such studies may have limited power, especially when susceptibility genes are rare and exposure distributions are highly skewed. We propose a variant of the classical case‐control study, the exposure enriched case‐control (EECC) design, where not only cases, but also high (or low) exposed individuals are oversampled, depending on the skewness of the exposure distribution. Of course, a traditional logistic regression model is no longer valid and results in biased parameter estimation. We show that addition of a simple covariate to the regression model removes this bias and yields reliable estimates of main and interaction effects of interest. We also discuss optimal design, showing that judicious oversampling of high/low exposed individuals can boost study power considerably. We illustrate our results using data from a study involving arsenic exposure and detoxification genes in Bangladesh.
Hussain, W, Hussain, FK, Hussain, OK & Chang, E 2016, 'Provider-Based Optimized Personalized Viable SLA (OPV-SLA) Framework to Prevent SLA Violation', The Computer Journal, vol. 59, no. 12, pp. 1760-1783.
View/Download from: Publisher's site
View description>>
Service level agreement (SLA) is an essential agreement formed between a consumer and a provider in business activities. The SLA defines the business terms, objectives, obligations and commitment of both parties to a business activity, and in cloud computing it also defines a consumer's request for both fixed and variable resources, due to the elastic and dynamic nature of the cloud-computing environment. Providers need to thoroughly analyze such variability when forming SLAs to ensure they commit to the agreements with consumers and at the same time make the best use of available resources and obtain maximum returns. They can achieve this by entering into viable SLAs with consumers. A consumer's profile becomes a key element in determining the consumer's reliability, as a consumer who has previous service violation history is more likely to violate future service agreements; hence, a provider can avoid forming SLAs with such consumers. In this paper, we propose a novel optimal SLA formation architecture from the provider's perspective, enabling the provider to consider a consumer's reliability in committing to the SLA. We classify existing consumers into three categories based on their reliability or trustworthiness value and use that knowledge to ascertain whether to accept a consumer request for resource allocation, and then to determine the extent of the allocation. Our proposed architecture helps the service provider to monitor the behavior of service consumers in the post-interaction time phase and to use that information to form viable SLAs in the pre-interaction time phase to minimize service violations and penalties.
Khuat, TT & Le, MH 2016, 'Optimizing parameters of software effort estimation models using directed artificial bee colony algorithm', Informatica (Slovenia), vol. 40, no. 4, pp. 427-436.
View description>>
Effective software effort estimation is one of the challenging tasks in software engineering. There have been various alternatives introduced to enhance the accuracy of predictions. In this respect, estimation approaches based on algorithmic models have been widely used. These models consider modeling software effort as a function of the size of the developed project. However, most approaches sharing a common thread of complex mathematical models face the difficulties in parameters calibration and tuning. This study proposes using a directed artificial bee colony algorithm in order to tune the values of model parameters based on past actual effort. The proposed methods were verified with NASA software dataset and the obtained results were compared to the existing models in other literature. The results indicated that our proposal has significantly improved the performance of the estimations.
Khuat, TT, Le, QC, Nguyen, BL & Le, MH 2016, 'Forecasting stock price using wavelet neural network optimized by directed Artificial Bee Colony algorithm', Journal of Telecommunications and Information Technology, vol. 2016, no. 2, pp. 43-52.
View description>>
Stock prediction with data mining techniques is one of the most important issues in finance. This field has attracted great scientific interest and has become a crucial research area to provide a more precise prediction process. This study proposes an integrated approach where Haar wavelet transform and Artificial Neural Network optimized by Directed Artificial Bee Colony algorithm are combined for the stock price prediction. The proposed approach was tested on the historical price data collected from Yahoo Finance with different companies. Furthermore, the prediction result was found satisfactorily enough as a guide for traders and investors in making qualitative decisions.
Lan, C, Chen, Q & Li, J 2016, 'Grouping miRNAs of similar functions via weighted information content of gene ontology', BMC Bioinformatics, vol. 17, no. S19, pp. 159-295.
View/Download from: Publisher's site
View description>>
BACKGROUND: Regulation mechanisms between miRNAs and genes are complicated. To accomplish a biological function, a miRNA may regulate multiple target genes, and similarly a target gene may be regulated by multiple miRNAs. Wet-lab knowledge of co-regulating miRNAs is limited. This work introduces a computational method to group miRNAs of similar functions to identify co-regulating miRNAsfrom a similarity matrix of miRNAs. RESULTS: We define a novel information content of gene ontology (GO) to measure similarity between two sets of GO graphs corresponding to the two sets of target genes of two miRNAs. This between-graph similarity is then transferred as a functional similarity between the two miRNAs. Our definition of the information content is based on the size of a GO term's descendants, but adjusted by a weight derived from its depth level and the GO relationships at its path to the root node or to the most informative common ancestor (MICA). Further, a self-tuning technique and the eigenvalues of the normalized Laplacian matrix are applied to determine the optimal parameters for the spectral clustering of the similarity matrix of the miRNAs. CONCLUSIONS: Experimental results demonstrate that our method has better clustering performance than the existing edge-based, node-based or hybrid methods. Our method has also demonstrated a novel usefulness for the function annotation of new miRNAs, as reported in the detailed case studies.
Li, D, He, X, Cao, L & Chen, H 2016, 'Permutation anonymization', Journal of Intelligent Information Systems, vol. 47, no. 3, pp. 427-445.
View/Download from: Publisher's site
View description>>
In data publishing, anonymization techniques have been designed to provide privacy protection. Anatomy is an important techniques for privacy preserving in data publication and attracts considerable attention in the literature. However, anatomy is fragile under background knowledge attack and the presence attack. In addition, anatomy can only be applied into limited applications. To overcome these drawbacks, we propose an improved version of anatomy: permutation anonymization, a new anonymization technique that is more effective than anatomy in privacy protection, and in the meanwhile is able to retain significantly more information in the microdata. We present the detail of the technique and build the underlying theory of the technique. Extensive experiments on real data are conducted, showing that our technique allows highly effective data analysis, while offering strong privacy guarantees.
Li, J, Fong, S, Siu, S, Mohammed, S, Fiaidhi, J & Wong, KKL 2016, 'WITHDRAWN: Improving classification of protein binders for virtual drug screening by novel swarm-based feature selection techniques', Computerized Medical Imaging and Graphics.
View/Download from: Publisher's site
View description>>
© 2016 Elsevier Ltd. Drug design involves classification of protein binding which is usually done in a computer simulation prior to extensive actual tests. Accurate classification of protein binding is essential but it is obstructed with a very challenging task of feature selection (FS) because there are too many potential features. Dorothea as a case of virtual screening in drug design, has 100,000 features that inflate to a very huge (of size 2100,000 possible candidate feature subsets to be selected) but very sparse search space. In this paper, this computational challenge is tackled by a new model of feature selection called Two-stage Swarm Search-FS (TSS-FS). The novelty of TSS-FS is the use of adaptive search space shrinking mechanism which is the first stage of the TSS-FS to reduce computing cost and increase classification accuracy. Reducing the very huge and sparse search space enables the swarm feature selection operate more efficiently. Results demonstrated in the paper confirms the efficacy of the new algorithms.
Li, J, Zhao, B, Deng, C & Xu, RYD 2016, 'Time Varying Metric Learning for visual tracking', Pattern Recognition Letters, vol. 80, pp. 157-164.
View/Download from: Publisher's site
Li, Y, Li, Y & Xu, G 2016, 'Protecting private geosocial networks against practical hybrid attacks with heterogeneous information', Neurocomputing, vol. 210, pp. 81-90.
View/Download from: Publisher's site
View description>>
© 2016 Elsevier B.V.GeoSocial Networks (GSNs) are becoming increasingly popular due to its power in providing high-performance and flexible service capabilities. More and more Internet users have accepted this innovative service model. However, even GSNs have great business value for data analysis by integrated with location information, it may seriously compromise users' privacy in publishing the GSN data. In this paper, we study the identity disclosure problem in publishing GSN data. We first discuss the attack problem by considering both the location-based and structure-based properties, as background knowledge, and then formalize two general models, named (k,m)-anonymity and (k,m,l)-anonymity Then we propose a complete solution to achieve (k,m)-anonymization and (k,m,l)-anonymization to prevent the released data from the above attacks above. We also take data utility into consideration by defining specific information loss metrics. It is validated by real-world data that the proposed methods can prevent GSN dataset from the attacks while retaining good utility.
Liu, Q, Song, J & Li, J 2016, 'Using contrast patterns between true complexes and random subgraphs in PPI networks to predict unknown protein complexes', Scientific Reports, vol. 6, no. 1, pp. 1-15.
View/Download from: Publisher's site
View description>>
AbstractMost protein complex detection methods utilize unsupervised techniques to cluster densely connected nodes in a protein-protein interaction (PPI) network, in spite of the fact that many true complexes are not dense subgraphs. Supervised methods have been proposed recently, but they do not answer why a group of proteins are predicted as a complex, and they have not investigated how to detect new complexes of one species by training the model on the PPI data of another species. We propose a novel supervised method to address these issues. The key idea is to discover emerging patterns (EPs), a type of contrast pattern, which can clearly distinguish true complexes from random subgraphs in a PPI network. An integrative score of EPs is defined to measure how likely a subgraph of proteins can form a complex. New complexes thus can grow from our seed proteins by iteratively updating this score. The performance of our method is tested on eight benchmark PPI datasets and compared with seven unsupervised methods, two supervised and one semi-supervised methods under five standards to assess the quality of the predicted complexes. The results show that in most cases our method achieved a better performance, sometimes significantly.
Llopis-Albert, C, Merigó, JM & Xu, Y 2016, 'A coupled stochastic inverse/sharp interface seawater intrusion approach for coastal aquifers under groundwater parameter uncertainty', Journal of Hydrology, vol. 540, pp. 774-783.
View/Download from: Publisher's site
View description>>
© 2016 Elsevier B.V. This paper presents an alternative approach to deal with seawater intrusion problems, that overcomes some of the limitations of previous works, by coupling the well-known SWI2 package for MODFLOW with a stochastic inverse model named GC method. On the one hand, the SWI2 allows a vertically integrated variable-density groundwater flow and seawater intrusion in coastal multi-aquifer systems, and a reduction in number of required model cells and the elimination of the need to solve the advective-dispersive transport equation, which leads to substantial model run-time savings. On the other hand, the GC method allows dealing with groundwater parameter uncertainty by constraining stochastic simulations to flow and mass transport data (i.e., hydraulic conductivity, freshwater heads, saltwater concentrations and travel times) and also to secondary information obtained from expert judgment or geophysical surveys, thus reducing uncertainty and increasing reliability in meeting the environmental standards. The methodology has been successfully applied to a transient movement of the freshwater-seawater interface in response to changing freshwater inflow in a two-aquifer coastal aquifer system, where an uncertainty assessment has been carried out by means of Monte Carlo simulation techniques. The approach also allows partially overcoming the neglected diffusion and dispersion processes after the conditioning process since the uncertainty is reduced and results are closer to available data.
Llopis-Albert, C, Palacios-Marqués, D & Merigó, JM 2016, 'Decision making under uncertainty in environmental projects using mathematical simulation modeling', Environmental Earth Sciences, vol. 75, no. 19.
View/Download from: Publisher's site
View description>>
© 2016, Springer-Verlag Berlin Heidelberg. In decision-making processes, reliability and risk aversion play a decisive role. The aim of this study is to perform an uncertainty assessment of the effects of future scenarios of sustainable groundwater pumping strategies on the quantitative and chemical status of an aquifer. The good status of the aquifer is defined according to the terms established by the EU Water Framework Directive (WFD). A decision support systems (DSS) is presented, which makes use of a stochastic inverse model (GC method) and geostatistical approaches to calibrate equally likely realizations of hydraulic conductivity (K) fields for a particular case study. These K fields are conditional to available field data, including hard and soft information. Then, different future scenarios of groundwater pumping strategies are generated, based on historical information and WFD standards, and simulated for each one of the equally likely K fields. The future scenarios lead to different environmental impacts and levels of socioeconomic development of the region and, hence, to a different degree of acceptance among stakeholders. We have identified the different stakeholders implied in the decision-making process, the objectives pursued and the alternative actions that should be considered by stakeholders in a public participation project (PPP). The MonteCarlo simulation provides a highly effective way for uncertainty assessment and allows presenting the results in a simple and understandable way even for non-experts stakeholders. The methodology has been successfully applied to a real case study and lays the foundations to perform a PPP and stakeholders’ involvement in a decision-making process as required by the WFD. The results of the methodology can help the decision-making process to come up with the best policies and regulations for a groundwater system under uncertainty in groundwater parameters and management strategies and involving stakeh...
Merigó, JM & Núñez, A 2016, 'Influential journals in health research: a bibliometric study', Globalization and Health, vol. 12, no. 1, p. 46.
View/Download from: Publisher's site
View description>>
Background
There is a wide range of intellectual work written about health research, which has been shaped by the evolution of diseases. This study aims to identify the leading journals over the last 25 years (1990-2014) according to a wide range of bibliometric indicators.
Methods
The study develops a bibliometric overview of all the journals that are currently indexed in Web of Science (WoS) database in any of the four categories connected to health research. The work classifies health research in nine subfields: Public Health, Environmental and Occupational Health, Health Management and Economics, Health Promotion and Health Behavior, Epidemiology, Health Policy and Services, Medicine, Health Informatics, Engineering and Technology, and Primary Care.
Results
The results indicate a wide dispersion between categories being the American Journal of Epidemiology, Environmental Health Perspectives, American Journal of Public Health, and Social Science & Medicine, the journals that have received the highest number of citations over the last 25 years. According to other indicators such as the h-index and the citations per paper, some other journals such as the Annual Review of Public Health and Medical Care, obtain better results which show the wide diversity and profiles of outlets available in the scientific community. The results are grouped and studied according to the nine subfields in order to identify the leading journals in each specific sub discipline of health.
Conclusions
The work identifies the leading journals in health research through a bibliometric approach. The analysis shows a deep overview of the results of health journals. It is worth noting that many journals have entered the WoS database during the last years, in many cases to fill some specific niche that has emerged in the literature, although the most popular ones have been in the database for a long time.
Merigó, JM, Cancino, CA, Coronado, F & Urbano, D 2016, 'Academic research in innovation: a country analysis', Scientometrics, vol. 108, no. 2, pp. 559-593.
View/Download from: Publisher's site
Merigó, JM, Gil-Lafuente, AM & Gil-Lafuente, J 2016, 'Business, industrial marketing and uncertainty', Journal of Business & Industrial Marketing, vol. 31, no. 3, pp. 325-327.
View/Download from: Publisher's site
View description>>
PurposeThis special issue of the Journal of Business & Industrial Marketing, entitled “Business, Industrial Marketing and Uncertainty”, presents selected extended studies that were presented at the European Academy of Management and Business Economics Conference (AEDEM 2012).Design/methodology/approachThe main focus of this year was reflected in the slogan: “Creating new opportunities in an uncertain environment”. The objective was to show the importance that uncertainty has in our current world, strongly affected by many complexities and modern developments, especially through the new technological advances.FindingsOne fundamental reason that explains the economic crisis is that the government and companies were not well prepared for these critical situations. And the main justification for this is that they did not have enough information. Otherwise, they would have tried any possible strategy to avoid the crisis. Usually, uncertainty is defined as the situation with unknown information in the environment.Originality/valueFrom a theoretical perspective, the problem here is that enterprises and governments should assess the information and the uncertainty in a more appropriate way. Usually, they have some studies in this direction, but many times, it is not enough, as it was proved in the last economic crisis.
Merigó, JM, Palacios-Marqués, D & Ribeiro-Navarrete, B 2016, 'Corrigendum to “Aggregation systems for sales forecasting” [J. Bus. Res. 68(11) (2015) 2299–2304]', Journal of Business Research, vol. 69, no. 6, pp. 2325-2325.
View/Download from: Publisher's site
Merigó, JM, Palacios-Marqués, D & Zeng, S 2016, 'Subjective and objective information in linguistic multi-criteria group decision making', European Journal of Operational Research, vol. 248, no. 2, pp. 522-531.
View/Download from: Publisher's site
View description>>
Linguistic decision making systems represent situations that cannot be assessed with numerical information but it is possible to use linguistic variables. This paper introduces new linguistic aggregation operators in order to develop more efficient decision making systems. The linguistic probabilistic weighted average (LPWA) is presented. Its main advantage is that it considers subjective and objective information in the same formulation and considering the degree of importance that each concept has in the aggregation. A key feature of the LPWA operator is that it considers a wide range of linguistic aggregation operators including the linguistic weighted average, the linguistic probabilistic aggregation and the linguistic average. Further generalizations are presented by using quasi-arithmetic means and moving averages. An application in linguistic multi-criteria group decision making under subjective and objective risk is also presented in the context of the European Union law.
Merigó, JM, Peris-Ortíz, M, Navarro-García, A & Rueda-Armengot, C 2016, 'Aggregation operators in economic growth analysis and entrepreneurial group decision-making', Applied Soft Computing, vol. 47, pp. 141-150.
View/Download from: Publisher's site
View description>>
© 2016 Elsevier B.V. All rights reserved. An economic crisis can be measured from different perspectives. A very commonly used measure is that of a country's economic growth. When growth is lower than desired, the economy is assumed to be near stagnation or in an economic recession. This paper connects entrepreneurship and economic growth in decision-making problems assessed with modern aggregation systems. Aggregation techniques can represent information more comprehensively in uncertain and imprecise environments. This paper suggests several practical aggregation operators for this purpose, such as the ordered weighted average and the probabilistic ordered weighted averaging weighted average. Other aggregation systems based on macroeconomic theory are also introduced. The paper concludes with an application in an entrepreneurial uncertain multi-criteria multi-person decision-making problem regarding the selection of optimal markets for creating a new company. This approach is based on the use of economic growth as the fundamental variable for determining the preferred solution.
MERIGÓ, JM, ROCAFORT, A & AZNAR-ALARCÓN, JP 2016, 'BIBLIOMETRIC OVERVIEW OF BUSINESS & ECONOMICS RESEARCH', Journal of Business Economics and Management, vol. 17, no. 3, pp. 397-413.
View/Download from: Publisher's site
View description>>
Bibliometrics is the quantitative study of bibliographic information. It classifies the information according to different criteria including authors, journals, institutions and countries. This paper presents a general bibliometric overview of the most influential research in business & economics according to the information found in the Web of Science. It includes research from different subcategories including business, business finance, economics and management. For doing so, four general lists are presented: the 50 most cited papers in business & economics of all time, the 40 most influential journals, the 40 most relevant institutions and the most influential countries. The results permit to obtain a general picture of the most significant research in business & economics. This information is very useful in order to identify the leading trends in this area.
Merigó, JM, Yang, J-B & Xu, D-L 2016, 'Demand Analysis with Aggregation Systems', International Journal of Intelligent Systems, vol. 31, no. 5, pp. 425-443.
View/Download from: Publisher's site
Nasrallah, R, Fast, EM, Solaimani, P, Knezevic, K, Eliades, A, Patel, R, Thambyrajah, R, Unnikrishnan, A, Thoms, J, Beck, D, Vink, CS, Smith, A, Wong, J, Shepherd, M, Kent, D, Roychoudhuri, R, Paul, F, Klippert, J, Hammes, A, Willnow, T, Göttgens, B, Dzierzak, E, Zon, LI, Lacaud, G, Kouskoff, V & Pimanda, JE 2016, 'Identification of novel regulators of developmental hematopoiesis using Endoglin regulatory elements as molecular probes', Blood, vol. 128, no. 15, pp. 1928-1939.
View/Download from: Publisher's site
View description>>
Key PointsENG regulatory elements target hemogenic mesoderm and hemogenic endothelium. Hemogenic progenitors can be enriched using these elements as molecular probes to discover novel regulators of hematopoiesis.
Nguyen, Q, Khalifa, N, Alzamora, P, Gleeson, A, Catchpoole, D, Kennedy, P & Simoff, S 2016, 'Visual Analytics of Complex Genomics Data to Guide Effective Treatment Decisions', Journal of Imaging, vol. 2, no. 4, pp. 29-29.
View/Download from: Publisher's site
View description>>
© 2016 by the authors. In cancer biology, genomics represents a big data problem that needs accurate visual data processing and analytics. The human genome is very complex with thousands of genes that contain the information about the individual patients and the biological mechanisms of their disease. Therefore, when building a framework for personalised treatment, the complexity of the genome must be captured in meaningful and actionable ways. This paper presents a novel visual analytics framework that enables effective analysis of large and complex genomics data. By providing interactive visualisations from the overview of the entire patient cohort to the detail view of individual genes, our work potentially guides effective treatment decisions for childhood cancer patients. The framework consists of multiple components enabling the complete analytics supporting personalised medicines, including similarity space construction, automated analysis, visualisation, gene-to-gene comparison and user-centric interaction and exploration based on feature selection. In addition to the traditional way to visualise data, we utilise the Unity3D platform for developing a smooth and interactive visual presentation of the information. This aims to provide better rendering, image quality, ergonomics and user experience to non-specialists or young users who are familiar with 3D gaming environments and interfaces. We illustrate the effectiveness of our approach through case studies with datasets from childhood cancers, B-cell Acute Lymphoblastic Leukaemia (ALL) and Rhabdomyosarcoma (RMS) patients, on how to guide the effective treatment decision in the cohort.
Peng, F, Lu, J, Wang, Y, Xu, RY-D, Ma, C & Yang, J 2016, 'N -dimensional Markov random field prior for cold-start recommendation', Neurocomputing, vol. 191, pp. 187-199.
View/Download from: Publisher's site
View description>>
© 2016 Elsevier B.V. A recommender system is a commonly used technique to improve user experience in e-commerce applications. One of the popular recommender methods is Matrix Factorization (MF) that learns the latent profile of both users and items. However, if the historical ratings are not available, the latent profile will draw from a zero-mean Gaussian prior, resulting in uninformative recommendations. To deal with this issue, we propose using an n-dimensional Markov random field as the prior of matrix factorization (called mrf-MF). In the Markov random field, the attribute (such as age, occupation of users and genre, release year of items) is considered as the site and the latent profile, the random variable. Through the prior, new users or items will be recommended according to its neighbors. The proposed model is suitable for three types of cold-start recommendation: (1) recommend new items to existing users; (2) recommend new users for existing items; (3) recommend new items to new users. The proposed model is assessed on two movie datasets, Movielens 100K and Movielens 1M. Experimental results show that it can effectively solve each of the three cold-start problems and outperforms several matrix factorization based methods.
Perera, D, Poulos, RC, Shah, A, Beck, D, Pimanda, JE & Wong, JWH 2016, 'Differential DNA repair underlies mutation hotspots at active promoters in cancer genomes', NATURE, vol. 532, no. 7598, pp. 259-+.
View/Download from: Publisher's site
View description>>
© 2016 Macmillan Publishers Limited. Promoters are DNA sequences that have an essential role in controlling gene expression. While recent whole cancer genome analyses have identified numerous hotspots of somatic point mutations within promoters, many have not yet been shown to perturb gene expression or drive cancer development. As such, positive selection alone may not adequately explain the frequency of promoter point mutations in cancer genomes. Here we show that increased mutation density at gene promoters can be linked to promoter activity and differential nucleotide excision repair (NER). By analysing 1,161 human cancer genomes across 14 cancer types, we find evidence for increased local density of somatic point mutations within the centres of DNase I-hypersensitive sites (DHSs) in gene promoters. Mutated DHSs were strongly associated with transcription initiation activity, in which active promoters but not enhancers of equal DNase I hypersensitivity were most mutated relative to their flanking regions. Notably, analysis of genome-wide maps of NER shows that NER is impaired within the DHS centre of active gene promoters, while XPC-deficient skin cancers do not show increased promoter mutation density, pinpointing differential NER as the underlying cause of these mutation hotspots. Consistent with this finding, we observe that melanomas with an ultraviolet-induced DNA damage mutation signature show greatest enrichment of promoter mutations, whereas cancers that are not highly dependent on NER, such as colon cancer, show no sign of such enrichment. Taken together, our analysis has uncovered the presence of a previously unknown mechanism linking transcription initiation and NER as a major contributor of somatic point mutation hotspots at active gene promoters in cancer genomes.
Richmond, J, Robbins, A, Evans, K, Beck, D, Kurmasheva, RT, Billups, CA, Carol, H, Heatley, S, Sutton, R, Marshall, GM, White, D, Pimanda, J, Houghton, PJ, Smith, MA & Lock, RB 2016, 'Acute Sensitivity of Ph-like Acute Lymphoblastic Leukemia to the SMAC-Mimetic Birinapant', Cancer Research, vol. 76, no. 15, pp. 4579-4591.
View/Download from: Publisher's site
View description>>
Abstract Ph-like acute lymphoblastic leukemia (ALL) is a genetically defined high-risk ALL subtype with a generally poor prognosis. In this study, we evaluated the efficacy of birinapant, a small-molecule mimetic of the apoptotic regulator SMAC, against a diverse set of ALL subtypes. Birinapant exhibited potent and selective cytotoxicity against B-cell precursor ALL (BCP-ALL) cells that were cultured ex vivo or in vivo as patient-derived tumor xenografts (PDX). Cytotoxicity was consistently most acute in Ph-like BCP-ALL. Unbiased gene expression analysis of BCP-ALL PDX specimens identified a 68-gene signature associated with birinapant sensitivity, including an enrichment for genes involved in inflammatory response, hematopoiesis, and cell death pathways. All Ph-like PDXs analyzed clustered within this 68-gene classifier. Mechanistically, birinapant sensitivity was associated with expression of TNF receptor TNFR1 and was abrogated by interfering with the TNFα/TNFR1 interaction. In combination therapy, birinapant enhanced the in vivo efficacy of an induction-type regimen of vincristine, dexamethasone, and L-asparaginase against Ph-like ALL xenografts, offering a preclinical rationale to further evaluate this SMAC mimetic for BCP-ALL treatment. Cancer Res; 76(15); 4579–91. ©2016 AACR.
Ryan, LM, Wand, MP & Malecki, AA 2016, 'Bringing Coals to Newcastle', Significance, vol. 13, no. 6, pp. 32-37.
View/Download from: Publisher's site
View description>>
Abstract Making effective public policy decisions is challenging at the best of times, but especially in the context of environmental regulation, which typically requires managing opposing interests and strong opinions from industry and private citizens. In this case study, Louise Ryan, Matt Wand and Alan Malecki show how statistical analysis can help resolve conflict and inform effective decision-making under uncertainty
Salvador, MM, Budka, M & Gabrys, B 2016, 'Effects of Change Propagation Resulting from Adaptive Preprocessing in Multicomponent Predictive Systems', Procedia Computer Science, vol. 96, pp. 713-722.
View/Download from: Publisher's site
View description>>
Predictive modelling is a complex process that requires a number of steps to transform raw data into predictions. Preprocessing of the input data is a key step in such process, and the selection of proper preprocessing methods is often a labour intensive task. Such methods are usually trained offline and their parameters remain fixed during the whole model deployment lifetime. However, preprocessing of non-stationary data streams is more challenging since the lack of adaptation of such preprocessing methods may degrade system performance. In addition, dependencies between different predictive system components make the adaptation process more challenging. In this paper we discuss the effects of change propagation resulting from using adaptive preprocessing in a Multicomponent Predictive System (MCPS). To highlight various issues we present four scenarios with different levels of adaptation. A number of experiments have been performed with a range of datasets to compare the prediction error in all four scenarios. Results show that well managed adaptation considerably improves the prediction performance. However, the model can become inconsistent if adaptation in one component is not correctly propagated throughout the rest of system components. Sometimes, such inconsistency may not cause an obvious deterioration in the system performance, therefore being difficult to detect. In some other cases it may even lead to a system failure as was observed in our experiments.
Shen, B, Cao, L, Yao, M & Gao, Y 2016, 'Mining preferred navigation patterns by consolidating both selection and time preferences', World Wide Web, vol. 19, no. 5, pp. 979-1007.
View/Download from: Publisher's site
View description>>
© 2015, Springer Science+Business Media New York. Preferred navigation patterns (PNP) are those contiguous sequential patterns whose elements are preferred by users to be selected as the next steps between several different selections and are preferred by users to spend much time on. Such navigation path and time preferred patterns are more actionable than any other finds only considering either path or time in various web applications, such as web user navigation, targeted online advertising and recommendation. However, due to the conceptual confusion and limitation on navigation preference in the existing work, the corresponding algorithms cannot discover actionable preferred navigation patterns. In this paper, we study the problem of preferred navigation pattern mining by involving both navigation path and time length. Firstly, we carefully define the concepts of time preference and selection preference for time-related path sequences, which can well reflect user interests from the relative path selection and time consumption respectively. Secondly, we propose an efficient PNP-forest algorithm for identifying PNPs, by first introducing PNP-forest data structure, and then presenting PNP-forest growth and maintenance mechanism, associated with optimization strategies. Then we introduce a more efficient mining algorithm called PrefixSpan_Forest, which integrates the advantages of PrefixSpan and PNP-forest. The performance of these two algorithms are also evaluated and the results show that the algorithms can discover PNPs effectively.
Shields, BJ, Jackson, JT, Metcalf, D, Shi, W, Huang, Q, Garnham, AL, Glaser, SP, Beck, D, Pimanda, JE, Bogue, CW, Smyth, GK, Alexander, WS & McCormack, MP 2016, 'Acute myeloid leukemia requires Hhex to enable PRC2-mediated epigenetic repression of Cdkn2a', Genes & Development, vol. 30, no. 1, pp. 78-91.
View/Download from: Publisher's site
View description>>
Unlike clustered HOX genes, the role of nonclustered homeobox gene family members in hematopoiesis and leukemogenesis has not been extensively studied. Here we found that the hematopoietically expressed homeobox gene Hhex is overexpressed in acute myeloid leukemia (AML) and is essential for the initiation and propagation of MLL-ENL-induced AML but dispensable for normal myelopoiesis, indicating a specific requirement for Hhex for leukemic growth. Loss of Hhex leads to expression of the Cdkn2a-encoded tumor suppressors p16INK4a and p19ARF, which are required for growth arrest and myeloid differentiation following Hhex deletion. Mechanistically, we show that Hhex binds to the Cdkn2a locus and directly interacts with the Polycomb-repressive complex 2 (PRC2) to enable H3K27me3-mediated epigenetic repression. Thus, Hhex is a potential therapeutic target that is specifically required for AML stem cells to repress tumor suppressor pathways and enable continued self-renewal.
Stoianoff, NP & Walpole, M 2016, 'Tax and the Environment: An Evaluation Framework for Tax Policy Reform — Group Delphi Study', Australian Tax Forum, vol. 31(4), no. 4, pp. 693-716.
View description>>
This paper reports on the Delphi Study undertaken by the authors in the development of a tax policy analysis framework that can be utilised to evaluate the effectiveness of Environmental Tax Measures (ETMs), building that framework from a critical assessment of the menu of factors advanced as possibilities in the prior literature. The ETMs are more commonly referred to as tax concessions or subsidies and are a form of tax expenditure used by governments to intervene in markets and influence the behaviour of particular taxpayers or industries. Such concessions need to be evaluated to assess their efficiency and effectiveness among other criteria. The Delphi Study was undertaken by bringing together an international group of expert environmental taxation scholars (the Reference Group) to participate in a Roundtable held during the 16th Global Conference on Environmental Taxation at UTS in September 2015. This is a variation on the Group Delphi method. While the Delphi method is traditionally based on anonymity, the Group Delphi assembles the expert panel to take part in a structured communication process using rotating subgroups to address the relevant questionnaire(s) (applying Likert scaling) and open questions, building consensus and defining disagreement by employing plenary discussions between iterations to foster peer review. The Roundtable utilised a single group
Sun, L, Ma, J, Zhang, Y, Dong, H & Hussain, FK 2016, 'Cloud-FuSeR: Fuzzy ontology and MCDM based cloud service selection', Future Generation Computer Systems, vol. 57, pp. 42-55.
View/Download from: Publisher's site
Tee, AE, Liu, B, Song, R, Li, J, Pasquier, E, Cheung, BB, Jiang, C, Marshall, GM, Haber, M, Norris, MD, Fletcher, JI, Dinger, ME & Liu, T 2016, 'The long noncoding RNA MALAT1 promotes tumor-driven angiogenesis by up-regulating pro-angiogenic gene expression', Oncotarget, vol. 7, no. 8, pp. 8663-8675.
View/Download from: Publisher's site
View description>>
Neuroblastoma is the most common solid tumor during early childhood. One of the key features of neuroblastoma is extensive tumor-driven angiogenesis due to hypoxia. However, the mechanism through which neuroblastoma cells drive angiogenesis is poorly understood. Here we show that the long noncoding RNA MALAT1 was upregulated in human neuroblastoma cell lines under hypoxic conditions. Conditioned media from neuroblastoma cells transfected with small interfering RNAs (siRNA) targeting MALAT1, compared with conditioned media from neuroblastoma cells transfected with control siRNAs, induced significantly less endothelial cell migration, invasion and vasculature formation. Microarray-based differential gene expression analysis showed that one of the genes most significantly down-regulated following MALAT1 suppression in human neuroblastoma cells under hypoxic conditions was fibroblast growth factor 2 (FGF2). RT-PCR and immunoblot analyses confirmed that MALAT1 suppression reduced FGF2 expression, and Enzyme-Linked Immunosorbent Assays revealed that transfection with MALAT1 siRNAs reduced FGF2 protein secretion from neuroblastoma cells. Importantly, addition of recombinant FGF2 protein to the cell culture media reversed the effects of MALAT1 siRNA on vasculature formation. Taken together, our data suggest that up-regulation of MALAT1 expression in human neuroblastoma cells under hypoxic conditions increases FGF2 expression and promotes vasculature formation, and therefore plays an important role in tumor-driven angiogenesis.
Valenzuela-Fernández, L, Nicolas, C, Gil-Lafuente, J & Merigó, JM 2016, 'Fuzzy indicators for customer retention', International Journal of Engineering Business Management, vol. 8, pp. 184797901667052-184797901667052.
View/Download from: Publisher's site
View description>>
It is widely known that market orientation (MO) and customer value help companies achieve sustainable sales growth over time. Nevertheless, one cannot ignore the existence of a gap on how to measure this relationship. Following this idea, this study proposes six fuzzy key performance indicators that aims to measure customer retention and loyalty of the portfolio. The work uses 300 sales executives. This exploratory study concludes that indicators such as MO, customer orientation (CO), degree of CO value of sales force, innovation capability, lifetime value, and customer service quality positively influence customer retention and loyalty portfolio.
Vaughan, N & Gabrys, B 2016, 'Comparing and Combining Time Series Trajectories Using Dynamic Time Warping', Procedia Computer Science, vol. 96, pp. 465-474.
View/Download from: Publisher's site
View description>>
This research proposes the application of dynamic time warping (DTW) algorithm to analyse multivariate data from virtual reality training simulators, to assess the skill level of trainees. We present results of DTW algorithm applied to trajectory data from a virtual reality haptic training simulator for epidural needle insertion. The proposed application of DTW algorithm serves two purposes, to enable (i) two trajectories to be compared as a similarity measure and also enables (ii) two or more trajectories to be combined together to produce a typical or representative average trajectory using a novel hierarchical DTW process. Our experiments included 100 expert and 100 novice simulator recordings. The data consists of multivariate time series data-streams including multi-dimensional trajectories combined with force and pressure measurements. Our results show that our proposed application of DTW provides a useful time-independent method for (i) comparing two trajectories by providing a similarity measure and (ii) combining two or more trajectories into one, showing higher performance compared to conventional methods such as linear mean. These results demonstrate that DTW can be useful within virtual reality training simulators to provide a component in an automated scoring and assessment feedback system.
Vaughan, N, Gabrys, B & Dubey, VN 2016, 'An overview of self-adaptive technologies within virtual reality training', Computer Science Review, vol. 22, pp. 65-87.
View/Download from: Publisher's site
View description>>
This overview presents the current state-of-the-art of self-adaptive technologies within virtual reality (VR) training. Virtual reality training and assessment is increasingly used for five key areas: medical, industrial & commercial training, serious games, rehabilitation and remote training such as Massive Open Online Courses (MOOCs). Adaptation can be applied to five core technologies of VR including haptic devices, stereo graphics, adaptive content, assessment and autonomous agents. Automation of VR training can contribute to automation of actual procedures including remote and robotic assisted surgery which reduces injury and improves accuracy of the procedure. Automated haptic interaction can enable tele-presence and virtual artefact tactile interaction from either remote or simulated environments. Automation, machine learning and data driven features play an important role in providing trainee-specific individual adaptive training content. Data from trainee assessment can form an input to autonomous systems for customised training and automated difficulty levels to match individual requirements. Self-adaptive technology has been developed previously within individual technologies of VR training. One of the conclusions of this research is that while it does not exist, an enhanced portable framework is needed and it would be beneficial to combine automation of core technologies, producing a reusable automation framework for VR training.
Waller, D, Mondy, P, Brama, T, Fisher, J, King, A, Malkov, K, Wall‐Smith, D, Ryan, L & Irving, DO 2016, 'Determining the effect of vein visualization technology on donation success, vasovagal symptoms, anxiety and intention to re‐donate in whole blood donors aged 18–30 years: A randomized controlled trial', Vox Sanguinis, vol. 111, no. 2, pp. 135-143.
View/Download from: Publisher's site
View description>>
Background and objectivesVein visualization technology (VVT) devices use near‐infrared light to assist location of peripheral veins. The current study investigated the impact of VVT on donor experience and collection success for young blood donors at the Australian Red Cross Blood Service.Materials and MethodsThe study in donors aged 18 to 30 years used a two intervention to one control randomized trial design with 285 new and 587 returning donors recruited at two sites. Donors reported presyncopal symptoms, phlebotomy pain, anxiety and intentions to redonate along with other measures. Participating phlebotomists rated usefulness of the technology. Flow rates, collection volumes and other donation information were taken from routine data.ResultsNo significant differences were found between control and intervention groups on presyncopal symptoms, phlebotomy pain, anxiety, intentions to redonate, flow rates, collection volumes or vasovagal reactions (all P's > 0·05). Phlebotomist ratings of VVT were significantly more positive when they had less than 5 years of experience (P < 0·01) or when the vein was not visible to the naked eye (P < 0·01).ConclusionsResults suggest that VVT does not improve the donation experience for younger blood donors. Staff reports indicate that VVT may have some utility for assisting with difficult phlebotomi...
Wang, C, Dong, X, Han, L, Su, X-D, Zhang, Z, Li, J & Song, J 2016, 'Identification of WD40 repeats by secondary structure-aided profile–profile alignment', Journal of Theoretical Biology, vol. 398, pp. 122-129.
View/Download from: Publisher's site
View description>>
A WD40 protein typically contains four or more repeats of ~40 residues ended with the Trp-Asp dipeptide, which folds into β-propellers with four β strands in each repeat. They often function as scaffolds for protein-protein interactions and are involved in numerous fundamental biological processes. Despite their important functional role, the "velcro" closure of WD40 propellers and the diversity of WD40 repeats make their identification a difficult task. Here we develop a new WD40 Repeat Recognition method (WDRR), which uses predicted secondary structure information to generate candidate repeat segments, and further employs a profile-profile alignment to identify the correct WD40 repeats from candidate segments. In particular, we design a novel alignment scoring function that combines dot product and BLOSUM62, thereby achieving a great balance of sensitivity and accuracy. Taking advantage of these strategies, WDRR could effectively reduce the false positive rate and accurately identify more remote homologous WD40 repeats with precise repeat boundaries. We further use WDRR to re-annotate the Pfam families in the β-propeller clan (CL0186) and identify a number of WD40 repeat proteins with high confidence across nine model organisms. The WDRR web server and the datasets are available at http://protein.cau.edu.cn/wdrr/.
Wang, C, Fang, Y & Li, J-Y 2016, 'stimation of NON-WSSUS Channel for OFDM System: Exploiting Support Correlations through a Novel Adaptive Weighted Predict-Re-Estimate L1 Minimization', Journal of Communications, vol. 11, no. 2, pp. 149-156.
View/Download from: Publisher's site
View description>>
© 2016 Journal of Communications. It is challenging to estimate the wireless channel of the Orthogonal Frequency-Division Multiplexing (OFDM) broadband system under a changing communication environment. The difficulty is mainly attributed to this wireless channel’s Non Wide Sense Stationary Uncorrelated Scattering (Non-WSSUS) which has an implication that the delay and Doppler shift of such a channel are non-stationary and correlated. A Non-WSSUS channel is very different from the classical time-varying channel with constant delay and Doppler shift. In this paper, we propose an estimation method for the Non-WSSUS Channel Impulse Response (CIR) of the OFDM system. Based on the sparsity property of the delay-Doppler spread function, the delay and Doppler shift of Non-WSSUS channel can be extracted through a Compressive Sensing (CS) approach. Then a novel CS algorithm referred as Pre-Re L1 is proposed. The proposed CS algorithm exploits the correlations of the sparse supports to obtain adaptive weights for L1minimization. Numerical Simulation results show that the proposed CS method improves the performance of the Non-WSSUS wireless channel estimation.
Wang, L, Patras, I, Zhang, J, Mori, G & Davis, L 2016, 'Special Issue on Individual and Group Activities in Video Event Analysis', Computer Vision and Image Understanding, vol. 144, pp. 1-2.
View/Download from: Publisher's site
Wang, W, Jiao, P, He, D, Jin, D, Pan, L & Gabrys, B 2016, 'Autonomous overlapping community detection in temporal networks: A dynamic Bayesian nonnegative matrix factorization approach', Knowledge-Based Systems, vol. 110, pp. 121-134.
View/Download from: Publisher's site
View description>>
A wide variety of natural or artificial systems can be modeled as time-varying or temporal networks. To understand the structural and functional properties of these time-varying networked systems, it is desirable to detect and analyze the evolving community structure. In temporal networks, the identified communities should reflect the current snapshot network, and at the same time be similar to the communities identified in history or say the previous snapshot networks. Most of the existing approaches assume that the number of communities is known or can be obtained by some heuristic methods. This is unsuitable and complicated for most real world networks, especially temporal networks. In this paper, we propose a Bayesian probabilistic model, named Dynamic Bayesian Nonnegative Matrix Factorization (DBNMF), for automatic detection of overlapping communities in temporal networks. Our model can not only give the overlapping community structure based on the probabilistic memberships of nodes in each snapshot network but also automatically determines the number of communities in each snapshot network based on automatic relevance determination. Thereafter, a gradient descent algorithm is proposed to optimize the objective function of our DBNMF model. The experimental results using both synthetic datasets and real-world temporal networks demonstrate that the DBNMF model has superior performance compared with two widely used methods, especially when the number of communities is unknown and when the network is highly sparse.
Wang, X, Yan, R, Li, J & Song, J 2016, 'SOHPRED: a new bioinformatics tool for the characterization and prediction of human S-sulfenylation sites', Molecular BioSystems, vol. 12, no. 9, pp. 2849-2858.
View/Download from: Publisher's site
View description>>
SOHPRED is a new and competitive bioinformatics tool for characterizing and predicting human S-sulfenylation sites.
Wang, Y, Zhang, J, Liu, Z, Wu, Q, Chou, PA, Zhang, Z & Jia, Y 2016, 'Handling Occlusion and Large Displacement Through Improved RGB-D Scene Flow Estimation', IEEE Transactions on Circuits and Systems for Video Technology, vol. 26, no. 7, pp. 1265-1278.
View/Download from: Publisher's site
View description>>
© 1991-2012 IEEE. The accuracy of scene flow is restricted by several challenges such as occlusion and large displacement motion. When occlusion happens, the positions inside the occluded regions lose their corresponding counterparts in preceding and succeeding frames. Large displacement motion will increase the complexity of motion modeling and computation. Moreover, occlusion and large displacement motion are highly related problems in scene flow estimation, e.g., large displacement motion often leads to considerably occluded regions in the scene. An improved dense scene flow method based on red-green-blue-depth (RGB-D) data is proposed in this paper. To handle occlusion, we model the occlusion status for each point in our problem formulation, and jointly estimate the scene flow and occluded regions. To deal with large displacement motion, we employ an over-parameterized scene flow representation to model both the rotation and translation components of the scene flow, since large displacement motion cannot be well approximated using translational motion only. Furthermore, we employ a two-stage optimization procedure for this overparameterized scene flow representation. In the first stage, we propose a new RGB-D PatchMatch method, which is mainly applied in the RGB-D image space to reduce the computational complexity introduced by the large displacement motion. According to the quantitative evaluation based on the Middlebury data set, our method outperforms other published methods. The improved performance is also comprehensively confirmed on the real data acquired by Kinect sensor.
Wu, J, Hong, Z, Pan, S, Zhu, X, Cai, Z & Zhang, C 2016, 'Multi-graph-view subgraph mining for graph classification', Knowledge and Information Systems, vol. 48, no. 1, pp. 29-54.
View/Download from: Publisher's site
View description>>
© 2015, Springer-Verlag London. In this paper, we formulate a new multi-graph-view learning task, where each object to be classified contains graphs from multiple graph-views. This problem setting is essentially different from traditional single-graph-view graph classification, where graphs are collected from one single-feature view. To solve the problem, we propose a cross graph-view subgraph feature-based learning algorithm that explores an optimal set of subgraphs, across multiple graph-views, as features to represent graphs. Specifically, we derive an evaluation criterion to estimate the discriminative power and redundancy of subgraph features across all views, with a branch-and-bound algorithm being proposed to prune subgraph search space. Because graph-views may complement each other and play different roles in a learning task, we assign each view with a weight value indicating its importance to the learning task and further use an optimization process to find optimal weight values for each graph-view. The iteration between cross graph-view subgraph scoring and graph-view weight updating forms a closed loop to find optimal subgraphs to represent graphs for multi-graph-view learning. Experiments and comparisons on real-world tasks demonstrate the algorithm’s superior performance.
Xia, F, Liu, H, Lee, I & Cao, L 2016, 'Scientific Article Recommendation: Exploiting Common Author Relations and Historical Preferences', IEEE Transactions on Big Data, vol. 2, no. 2, pp. 101-112.
View/Download from: Publisher's site
Ye, L, Liu, Z, Zhou, X, Shen, L & Zhang, J 2016, 'Saliency Detection Via Similar Image Retrieval', IEEE Signal Processing Letters, vol. 23, no. 6, pp. 838-842.
View/Download from: Publisher's site
Yi, X, Paulet, R, Bertino, E & Xu, G 2016, 'Private Cell Retrieval From Data Warehouses', IEEE Transactions on Information Forensics and Security, vol. 11, no. 6, pp. 1346-1361.
View/Download from: Publisher's site
View description>>
© 2015 IEEE. Publicly accessible data warehouses are an indispensable resource for data analysis. However, they also pose a significant risk to the privacy of the clients, since a data warehouse operator may follow the client's queries and infer what the client is interested in. Private information retrieval (PIR) techniques allow the client to retrieve a cell from a data warehouse without revealing to the operator which cell is retrieved and, therefore, protects the privacy of the client's queries. However, PIR cannot be used to hide online analytical processing (OLAP) operations performed by the client, which may disclose the client's interest. This paper presents a solution for private cell retrieval from a data warehouse on the basis of the Paillier cryptosystem. By our solution, the client can privately perform OLAP operations on the data warehouse and retrieve one (or more) cell without revealing any information about which cell is selected. In addition, we propose a solution for private block download on the basis of the Paillier cryptosystem. Our private block download allows the client to download an encrypted block from a data warehouse without revealing which block in a cloaking region is downloaded and improves the feasibility of our private cell retrieval. Our solutions ensure both the server's privacy and the client's privacy. Our experiments have shown that our solutions are practical.
Yu, D, Li, D-F & Merigó, JM 2016, 'Dual hesitant fuzzy group decision making method and its application to supplier selection', International Journal of Machine Learning and Cybernetics, vol. 7, no. 5, pp. 819-831.
View/Download from: Publisher's site
View description>>
The concept of dual hesitant fuzzy set arising from hesitant fuzzy set is generalized by including a function reflecting the decision maker’s fuzziness about the non-membership degree of the information provided. This paper studies some dual hesitant fuzzy information aggregation operators for aggregating dual hesitant fuzzy elements, such as dual hesitant fuzzy Heronian mean operator and dual hesitant fuzzy geometric Heronian mean operator. The research resulting dual hesitant fuzzy information aggregation operators finds an important role in group decision making (GDM) applications. It can fusion the experts’ opinion to the comprehensive ones and based on which an optimal decision making scheme can be determined. The properties of the proposed operators are studied and the application on GDM are investigated. The effectiveness of the GDM method is demonstrated on the case study about supplier selection.
Yu, D, Li, D-F, Merigó, JM & Fang, L 2016, 'Mapping development of linguistic decision making studies', Journal of Intelligent & Fuzzy Systems, vol. 30, no. 5, pp. 2727-2736.
View/Download from: Publisher's site
Yu, D, Merigó, JM & Xu, Y 2016, 'Group Decision Making in Information Systems Security Assessment Using Dual Hesitant Fuzzy Set', International Journal of Intelligent Systems, vol. 31, no. 8, pp. 786-812.
View/Download from: Publisher's site
View description>>
Network information system security has become a global issue since it is related to the economic development and national security. Information system security assessment plays an important role in the development of security solutions. Aiming at this issue, a dual hesitant fuzzy (DHF) group decision-making (GDM) method was proposed in this paper to assist the assessment of network information system security. A systemic index containing four aspects was established including organization security, management security, technical security, and personnel management security. The DHF group evaluation matrix was constructed based on the individual evaluation information from each expert. Some power average operator-based DHF information aggregation operators are proposed and used to fusion the performance of each criterion for information systems. The advantage of these operators is that they can describe the relationship between the indexes quantitatively. Finally, a case study about information systems security assessment was presented to verify the effectiveness of proposed GDM methods.
Zhang, J, Wu, Q, Shen, C, Zhang, J & Lu, J 2016, 'Multi-Label Image Classification with Regional Latent Semantic Dependencies', IEEE Transactions on Multimedia, vol. 20, no. 10, pp. 2801-2813.
View/Download from: Publisher's site
View description>>
Deep convolution neural networks (CNN) have demonstrated advanced performanceon single-label image classification, and various progress also have been madeto apply CNN methods on multi-label image classification, which requires toannotate objects, attributes, scene categories etc. in a single shot. Recentstate-of-the-art approaches to multi-label image classification exploit thelabel dependencies in an image, at global level, largely improving the labelingcapacity. However, predicting small objects and visual concepts is stillchallenging due to the limited discrimination of the global visual features. Inthis paper, we propose a Regional Latent Semantic Dependencies model (RLSD) toaddress this problem. The utilized model includes a fully convolutionallocalization architecture to localize the regions that may contain multiplehighly-dependent labels. The localized regions are further sent to therecurrent neural networks (RNN) to characterize the latent semanticdependencies at the regional level. Experimental results on several benchmarkdatasets show that our proposed model achieves the best performance compared tothe state-of-the-art models, especially for predicting small objects occurredin the images. In addition, we set up an upper bound model (RLSD+ft-RPN) usingbounding box coordinates during training, the experimental results also showthat our RLSD can approach the upper bound without using the bounding-boxannotations, which is more realistic in the real world.
Zhang, P, He, J, Long, G, Huang, G & Zhang, C 2016, 'Towards Anomalous Diffusion Sources Detection in a Large Network', ACM Transactions on Internet Technology, vol. 16, no. 1, pp. 1-24.
View/Download from: Publisher's site
View description>>
Witnessing the wide spread of malicious information in large networks, we develop an efficient method to detect anomalous diffusion sources and thus protect networks from security and privacy attacks. To date, most existing work on diffusion sources detection are based on the assumption that network snapshots that reflect information diffusion can be obtained continuously. However, obtaining snapshots of an entire network needs to deploy detectors on all network nodes and thus is very expensive. Alternatively, in this article, we study the diffusion sources locating problem by learning from information diffusion data collected from only a small subset of network nodes. Specifically, we present a new regression learning model that can detect anomalous diffusion sources by jointly solving five challenges, that is, unknown number of source nodes, few activated detectors, unknown initial propagation time, uncertain propagation path and uncertain propagation time delay. We theoretically analyze the strength of the model and derive performance bounds. We empirically test and compare the model using both synthetic and real-world networks to demonstrate its performance.
Zhang, Q, Zhang, P, Long, G, Ding, W, Zhang, C & Wu, X 2016, 'Online Learning from Trapezoidal Data Streams', IEEE Transactions on Knowledge and Data Engineering, vol. 28, no. 10, pp. 2709-2723.
View/Download from: Publisher's site
View description>>
© 1989-2012 IEEE. In this paper, we study a new problem of continuous learning from doubly-streaming data where both data volume and feature space increase over time. We refer to the doubly-streaming data as trapezoidal data streams and the corresponding learning problem as online learning from trapezoidal data streams. The problem is challenging because both data volume and data dimension increase over time, and existing online learning [1] , [2] , online feature selection [3] , and streaming feature selection algorithms [4] , [5] are inapplicable. We propose a new Online Learning with Streaming Features algorithm (OLSF for short) and its two variants, which combine online learning [1] , [2] and streaming feature selection [4] , [5] to enable learning from trapezoidal data streams with infinite training instances and features. When a new training instance carrying new features arrives, a classifier updates the existing features by following the passive-aggressive update rule [2] and updates the new features by following the structural risk minimization principle. Feature sparsity is then introduced by using the projected truncation technique. We derive performance bounds of the OL SF algorithm and its variants. We also conduct experiments on real-world data sets to show the performance of the proposed algorithms.
Zhang, Y, Wu, J, Cai, Z, Zhang, P & Chen, L 2016, 'Memetic Extreme Learning Machine', Pattern Recognition, vol. 58, pp. 135-148.
View/Download from: Publisher's site
View description>>
© 2016. Extreme Learning Machine (ELM) is a promising model for training single-hidden layer feedforward networks (SLFNs) and has been widely used for classification. However, ELM faces the challenge of arbitrarily selected parameters, e.g., the network weights and hidden biases. Therefore, many efforts have been made to enhance the performance of ELM, such as using evolutionary algorithms to explore promising areas of the solution space. Although evolutionary algorithms can explore promising areas of the solution space, they are not able to locate global optimum efficiently. In this paper, we present a new Memetic Algorithm (MA)-based Extreme Learning Machine (M-ELM for short). M-ELM embeds the local search strategy into the global optimization framework to obtain optimal network parameters. Experiments and comparisons on 46 UCI data sets validate the performance of M-ELM. The corresponding results demonstrate that M-ELM significantly outperforms state-of-the-art ELM algorithms.
Zhang, Z, Liu, Y, Xu, G & Chen, H 2016, 'A weighted adaptation method on learning user preference profile', Knowledge-Based Systems, vol. 112, pp. 114-126.
View/Download from: Publisher's site
View description>>
© 2016 Elsevier B.V. Recommender systems typically store personal preference profiles. Many items in the profiles can be represented by numerical attributes. However, the initial profile of each user is incomplete and imprecise. One important problem in the development of these systems is how to learn user preferences, and how to automatically adapted update the profiles. To address this issue, this paper presents an unsupervised approach for learning user preferences over numeric attributes by analyzing the interactions between users and recommender systems. When a list of recommendations shown to a target user, the favorite item will be selected by him/her, then the selected item and the over-ranked items will be employed as valuable feedback to learn the user profile. Specifically, two contributions are offered: 1), a learning approach to measure the influence of over-ranked items through analysis of user feedbacks and 2), a weighting algorithm to calculate weights of different attributes by analyzing user selections. These two approaches are integrated into a traditional adaption model for updating user preference profile. Extensive simulations and results show that both approaches are more effective than existing approaches.
Zhang, Z, Liu, Y, Xu, G & Luo, G 2016, 'Recommendation using DMF-based fine tuning method', Journal of Intelligent Information Systems, vol. 47, no. 2, pp. 233-246.
View/Download from: Publisher's site
View description>>
© 2016 Springer Science+Business Media New York Recommender Systems (RS) have been comprehensively analyzed in the past decade, Matrix Factorization (MF)-based Collaborative Filtering (CF) method has been proved to be an useful model to improve the performance of recommendation. Factors that inferred from item rating patterns shows the vectors which are useful for MF to characterize both items and users. A recommendation can concluded from good correspondence between item and user factors. A basic MF model starts with an object function, which is consisted of the squared error between original training matrix and predicted matrix as well as the regularization term (regularization parameters). To learn the predicted matrix, recommender systems minimize the squared error which has been regularized. However, two important details have been ignored: (1) the predicted matrix will be more and more accuracy as the iterations carried out, then a fix value of regularization parameters may not be the most suitable choice. (2) the final distribution trend of ratings of predicted matrix is not similar with the original training matrix. Therefore, we propose a Dynamic-MF algorithm and fine tuning method which is quite general to overcome the mentioned detail problems. Some other information, such as social relations, etc, can be easily incorporated into this method (model). The experimental analysis on two large datasets demonstrates that our approaches outperform the basic MF-based method.
Zhou, L, Merigó, JM, Chen, H & Liu, J 2016, 'The optimal group continuous logarithm compatibility measure for interval multiplicative preference relations based on the COWGA operator', Information Sciences, vol. 328, pp. 250-269.
View/Download from: Publisher's site
View description>>
The calculation of compatibility measures is an important technique employed in group decision-making with interval multiplicative preference relations. In this paper, a new compatibility measure called the continuous logarithm compatibility, which considers risk attitudes in decision-making based on the continuous ordered weighted geometric averaging (COWGA) operator, is introduced. We also develop a group continuous compatibility model (GCC Model) by minimizing the group continuous logarithm compatibility measure between the synthetic interval multiplicative preference relation and the continuous characteristic preference relation. Furthermore, theoretical foundations are established for the proposed model, such as the sufficient and necessary conditions for the existence of an optimal solution, the conditions for the existence of a superior optimal solution and the conditions for the existence of redundant preference relations. In addition, we investigate certain conditions for which the optimal objective function of the GCC Model guarantees its efficiency as the number of decision-makers increases. Finally, practical illustrative examples are examined to demonstrate the model and compare it with previous methods.
Abdullaev, S, McBurney, P & Musial, K 1970, 'Pricing options with portfolio-holding trading agents in direct double auction', Frontiers in Artificial Intelligence and Applications, 22nd European Conference on Artificial Intelligence (ECAI), IOS PRESS, Hague, NETHERLANDS, pp. 1754-1755.
View/Download from: Publisher's site
View description>>
Options constitute integral part of modern financial trades, and are priced according to the risk associated with buying or selling certain asset in future. Financial literature mostly concentrates on risk-neutral methods of pricing options such as Black-Scholes model. However, it is an emerging field in option pricing theory to use trading agents with utility functions to determine the option's potential payoff for the agent. In this paper, we use one of such methodologies developed by Othman and Sandholm to design portfolio-holding agents that are endowed with popular option portfolios such as bullish spread, butterfly spread, straddle, etc to price options. Agents use their portfolios to evaluate how buying or selling certain option would change their current payoff structure, and form their orders based on this information. We also simulate these agents in a multi-unit direct double auction. The emerging prices are compared to risk-neutral prices under different market conditions. Through an appropriate endowment of option portfolios to agents, we can also mimic market conditions where the population of agents are bearish, bullish, neutral or non-neutral in their beliefs.
Abidi, S, Piccardi, M & Williams, M-A 1970, 'Static action recognition by efficient greedy inference', 2016 IEEE Winter Conference on Applications of Computer Vision (WACV), 2016 IEEE Winter Conference on Applications of Computer Vision (WACV), IEEE, Lake Placid, NY, USA, pp. 1-8.
View/Download from: Publisher's site
View description>>
© 2016 IEEE. Action recognition from a single image is an important task for applications such as image annotation, robotic navigation, video surveillance and several others. Existing methods for recognizing actions from still images mainly rely on either bag-of-feature representations or pose estimation from articulated body-part models. However, the relationship between the action and the containing image is still substantially unexplored. Actually, the presence of given objects or specific backgrounds is likely to provide informative clues for the recognition of the action. For this reason, in this paper we propose approaching action recognition by first partitioning the entire image into superpixels, and then using their latent classes as attributes of the action. The action class is predicted based on a graphical model composed of measurements from each superpixel and a fully-connected graph of superpixel classes. The model is learned using a latent structural SVM approach, and an efficient, greedy algorithm is proposed to provide inference over the graph. Differently from most existing methods, the proposed approach does not require annotation of the actor (usually provided as a bounding box). Experimental results over the challenging Stanford 40 Action dataset have reported an impressive mean average precision of 72.3%, the highest achieved to date.
Alfaro-Garcia, VG, Gil-Lafuente, AM & Merigo, JM 1970, 'Induced generalized ordered weighted logarithmic aggregation operators', 2016 IEEE Symposium Series on Computational Intelligence (SSCI), 2016 IEEE Symposium Series on Computational Intelligence (SSCI), IEEE, Athens, Greexe, pp. 1-7.
View/Download from: Publisher's site
View description>>
© 2016 IEEE. We present the induced generalized ordered weighted logarithmic aggregation (IGOWLA) operator. It is an extension of the generalized ordered weighted logarithmic aggregation (GOWLA) operator. The IGOWLA operator uses order-induced variables that modify the reordering mechanism of the arguments to be aggregated. The main advantage of the induced process is the consideration of the complex attitude of the decision makers. We study some properties of the IGOWLA operator, such as idempotency, commutativity, boundedness and monotonicity. Finally we present an illustrative example of a group decision-making procedure using a multi-person analysis and the IGOWLA operator in the area of innovation management.
Alkalbani, AM, Ghamry, AM, Hussain, FK & Hussain, OK 1970, 'Predicting the sentiment of SaaS online reviews using supervised machine learning techniques', 2016 International Joint Conference on Neural Networks (IJCNN), 2016 International Joint Conference on Neural Networks (IJCNN), IEEE, Vancouver, CANADA, pp. 1547-1553.
View/Download from: Publisher's site
View description>>
© 2016 IEEE.There has been a dramatic increase in the sharing of opinions and information across different web platforms and social media, especially online product reviews. Cloud web portals, such as getApp.com, were designed to amalgamate cloud service information and to also examine how consumers evaluate their experience of using cloud computing products. The current literature shows the growing importance of online users' reviews, hence this study focuses on investigating consumers' feedback on Software-as-a-Service (SaaS) products by developing models to predict reviewers' attitudes. The goal of this paper is to develop prediction models to predict the sentiment of SaaS consumers' reviews (positive or negative). This research proposes five models that are based on five algorithms, the Support Vector Machine algorithm, Naive Bayes algorithm, Naive Bayes (Kernel) algorithm, k-nearest neighbors algorithm, and the decision tree algorithm to predict the attitude of SaaS reviews. The prediction accuracy of the space vector algorithm (5-fold cross-validation) is 92.37% which suggests that this algorithm is able to better determine the sentiment of online reviews compared with the other models. The results of this study provide valuable insight into online SaaS reviews and will assist in the design of SaaS review websites.
Alkalbani, AM, Ghamry, AM, Hussain, FK & Hussain, OK 1970, 'Sentiment Analysis and classification for Software as a Service Reviews', IEEE 30TH INTERNATIONAL CONFERENCE ON ADVANCED INFORMATION NETWORKING AND APPLICATIONS IEEE AINA 2016, International Conference on Advanced Information Networking and Applications (was ICOIN), IEEE, Crans-Montana, Switzerland, pp. 53-58.
View/Download from: Publisher's site
View description>>
© 2016 IEEE.With the rapid growth of cloud services, there has been a significant increase in the number of online consumer reviews and opinions on these services on different social media platforms. These reviews are a source of valuable information in regard to cloud market position and cloud consumer satisfaction. This study explores cloud consumers' reviews that reflect the user's experience with Software as a Service (SaaS) applications. The reviews were collected from different web portals, and around 4000 online reviews were analysed using sentiment analysis to identify the polarity of each review, that is, whether the sentiment being expressed is positive, negative, or neutral. Also, this research develops a model for predicting the sentiment of Software as a Service consumers' reviews using a supervised learning machine called a support vector machine (SVM). The sentiment results show that 62% of the reviews are positive which indicates that consumers are most likely satisfied with SaaS services. The results show that the prediction accuracy of the SVM-based Binary Occurrence approach (3-fold crossvalidation testing) is 92.30%, indicating it performs better in determining sentiment compared with other approaches (Term Occurrences, TFIDF). This work also provides valuable insight into online SaaS reviews and offers the research community the first SaaS polarity dataset.
Alkalbani, AM, Hussain, FK & IEEE 1970, 'A Comparative Study and Future Research Directions in Cloud Service Discovery', PROCEEDINGS OF THE 2016 IEEE 11TH CONFERENCE ON INDUSTRIAL ELECTRONICS AND APPLICATIONS (ICIEA), IEEE Conference on Industrial Electronics and Applications, IEEE, Dearborn, MI, United States, pp. 1049-1056.
View/Download from: Publisher's site
View description>>
© 2016 IEEE.Cloud computing technology is a new paradigm which provides Information Technology (IT) resources via the Internet. This new shift in the way that IT re-sources are offered to the user brings new challenges, such as cloud service discovery. Nowadays, cloud users are faced with a dilemma as they have an abundant choice of cloud services. Moreover, many cloud providers offer a range of services which deliver similar functionality. Locating the best and most appropriate cloud service with a suitable and capable provider is a primary concern for any consumer. In order to clearly comprehend the scope of this problem, a thorough analysis of the limitations of cloud service discovery approaches is required which, in turn, will empower researchers to deliver better solutions for consumers to make an informed decision and choose the right service. This paper presents an overview of the current cloud service discovery trends and challenges in recent studies. Additionally, the reviewed approaches are classified according to service discovery architecture and techniques. Furthermore, these approaches are compared and analysed from several perspectives including approach model/architecture, service type, ontology representation (domain, language, and reasoning), dynamic discovery model, evaluation model, user's preferences techniques, data updates, and public repositories.
Arellano, LAP, Castro, EL, Ochoa, EA & MerigoLindahl, JM 1970, 'Prioritized induced probabilistic OWA for dispute resolution methods', 2016 Annual Conference of the North American Fuzzy Information Processing Society (NAFIPS), 2016 Annual Conference of the North American Fuzzy Information Processing Society (NAFIPS), IEEE, Univ Texas El Paso, El Paso, TX.
View/Download from: Publisher's site
Awwad, S & Piccardi, M 1970, 'Local depth patterns for fine-grained activity recognition in depth videos', 2016 International Conference on Image and Vision Computing New Zealand (IVCNZ), 2016 International Conference on Image and Vision Computing New Zealand (IVCNZ), IEEE, Palmerston North, New Zealand, pp. 214-219.
View/Download from: Publisher's site
View description>>
© 2016 IEEE.Fine-grained activities are human activities involving small objects and small movements. Automatic recognition of such activities can prove useful for many applications, including detailed diarization of meetings and training sessions, assistive human-computer interaction and robotics interfaces. Existing approaches to fine-grained activity recognition typically leverage the combined use of multiple sensors including cameras, RFID tags, gyroscopes and accelerometers borne by the monitored people and target objects. Although effective, the downside of these solutions is that they require minute instrumentation of the environment that is intrusive and hard to scale. To this end, this paper investigates fine-grained activity recognition in a kitchen setting by solely using a depth camera. The primary contribution of this work is an aggregated depth descriptor that effectively captures the shape of the objects and the actors. Experimental results over the challenging '50 Salads' dataset of kitchen activities show an accuracy comparable to that of a state-of-the-art approach based on multiple sensors, thereby validating a less intrusive and more practical way of monitoring fine-grained activities.
Bakirov, R, Gabrys, B & Fay, D 1970, 'Augmenting adaptation with retrospective model correction for non-stationary regression problems', 2016 International Joint Conference on Neural Networks (IJCNN), 2016 International Joint Conference on Neural Networks (IJCNN), IEEE, Vancouver, CANADA, pp. 771-779.
View/Download from: Publisher's site
View description>>
Existing adaptive predictive methods often use multiple adaptive mechanisms as part of their coping strategy in non-stationary environments. We address a scenario when selective deployment of these adaptive mechanisms is possible. In this case, deploying each adaptive mechanism results in different candidate models, and only one of these candidates is chosen to make predictions on the subsequent data. After observing the error of each of candidate, it is possible to revert the current model to the one which had the least error. We call this strategy retrospective model correction. In this work we aim to investigate the benefits of such approach. As a vehicle for the investigation we use an adaptive ensemble method for regression in batch learning mode which employs several adaptive mechanisms to react to changes in the data. Using real world data from the process industry we show empirically that the retrospective model correction is indeed beneficial for the predictive accuracy, especially for the weaker adaptive mechanisms.
Beck, D, Thoms, J, Palu, C, Herold, T, Shah, A, Olivier, J, Boelen, L, Huang, Y, Chacon, D, Brown, A, Babic, M, Hahn, C, Perugini, M, Zhou, X, Huntly, B, Berdel, W, Woermann, B, Buechner, T, Hiddemann, W, Bohlander, S, Scott, H, Lewis, I, D'Andrea, R, Wong, J & Pimanda, J 1970, 'INTEGRATIVE ANALYSIS OF LINCRNA EXPRESSION IN 922 ACUTE MYELOID LEUKEMIA PATIENTS REVEALS MULTIPLE PROGNOSTIC GENE SIGNATURES', HAEMATOLOGICA, pp. 208-209.
Blanco-Mesa, F & Merigó, JM 1970, 'Bonferroni Means with the Adequacy Coefficient and the Index of Maximum and Minimum Level', Lecture Notes in Business Information Processing, Springer International Publishing, pp. 155-166.
View/Download from: Publisher's site
View description>>
© Springer International Publishing Switzerland 2016. The aim of the paper is to develop new aggregation operators using Bonferroni means, OWA operators and some distance and norms measures. We introduce the BON-OWAAC and BON-OWAIMAM operators. We are able to include adequacy coefficient and the maximum and minimum level in the same formulation with Bonferroni means and OWA operator. The main advantages on using these operators are that they allow considering continuous aggregations, multiple-comparison between each argument and distance measures in the same formulation. The numerical sample is focused on an entrepreneurial example in the sport industry in Colombia.
Blanco-Mesa, F & Merigo-Lindahl, JM 1970, 'Bonferroni distances with OWA operators', 2016 Annual Conference of the North American Fuzzy Information Processing Society (NAFIPS), 2016 Annual Conference of the North American Fuzzy Information Processing Society (NAFIPS), IEEE, El Paso, TX, USA.
View/Download from: Publisher's site
View description>>
© 2016 IEEE. The aim of the paper is to develop new aggregation operators using Bonferroni means, ordered weighted averaging (OWA) operators and some distance measures. We introduce the Bonferroni-Hamming weighted distance, Bonferroni OWA distance, and Bonferroni distances with OWA operators and weighted averages. The main advantages of using these operators are that they allow considering different aggregations contexts, multiple-comparison between each argument and distance measures in the same formulation.
Blanco-Mesa, F, Merigo Lindahl, JM & Gil-Lafuente, AM 1970, 'A bibliometric analysis of fuzzy decision making research', 2016 Annual Conference of the North American Fuzzy Information Processing Society (NAFIPS), 2016 Annual Conference of the North American Fuzzy Information Processing Society (NAFIPS), IEEE, El Paso, TX, USA.
View/Download from: Publisher's site
View description>>
© 2016 IEEE. Fuzzy decision-making consists in making decisions under complex and uncertain environments where the information can be assessed with fuzzy sets and systems. The aim of this study is to review the main contributions in this field by using a bibliometric approach. For doing so, the article uses a wide range of bibliometric indicators including the citations and the h-index. Moreover, it also uses the VOS viewer software in order to map the main trends in this area. The work considers the leading journals, articles, authors, institutions and countries. The results indicate that the Zadeh L.A. led the origins of fuzzy research and Ronald Yager is the most prominent author in FDM. The USA was the traditional leader in this field with the most significant researcher. However, during the last years, this field is receiving more attention by Asian authors that are starting to lead the field. This discipline has a strong potential and the expectations for the future is that it will continue to grow.
Braytee, A, Catchpoole, DR, Kennedy, PJ & Liu, W 1970, 'Balanced Supervised Non-Negative Matrix Factorization for Childhood Leukaemia Patients', Proceedings of the 25th ACM International on Conference on Information and Knowledge Management, CIKM'16: ACM Conference on Information and Knowledge Management, ACM, Indianapolis, Indiana, USA, pp. 2405-2408.
View/Download from: Publisher's site
View description>>
© 2016 ACM. Supervised feature extraction methods have received considerable attention in the data mining community due to their capability to improve the classification performance of the unsupervised dimensionality reduction methods. With increasing dimensionality, several methods based on supervised feature extraction are proposed to achieve a feature ranking especially on microarray gene expression data. This paper proposes a method with twofold objectives: it implements a balanced supervised non-negative matrix factorization (BSNMF) to handle the class imbalance problem in supervised non-negative matrix factorization techniques. Furthermore, it proposes an accurate gene ranking method based on our proposed BSNMF for microarray gene expression datasets. To the best of our knowledge, this is the first work to handle the class imbalance problem in supervised feature extraction methods. This work is part of a Human Genome project at The Children's Hospital at Westmead (TB-CHW), Australia. Our experiments indicate that the factorized components using supervised feature extraction approach have more classification capability than the unsu-pervised one, but it drastically fails at the presence of class imbalance problem. Our proposed method outperforms the state-of-the-art methods and shows promise in overcoming this concern.
Braytee, A, Liu, W & Kennedy, P 1970, 'A Cost-Sensitive Learning Strategy for Feature Extraction from Imbalanced Data', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), International Conference on Neural Information Processing, Springer International Publishing, Kyoto, Japan, pp. 78-86.
View/Download from: Publisher's site
View description>>
© Springer International Publishing AG 2016. In this paper, novel cost-sensitive principal component analysis (CSPCA) and cost-sensitive non-negative matrix factorization (CSNMF) methods are proposed for handling the problem of feature extraction from imbalanced data. The presence of highly imbalanced data misleads existing feature extraction techniques to produce biased features, which results in poor classification performance especially for the minor class problem. To solve this problem, we propose a costsensitive learning strategy for feature extraction techniques that uses the imbalance ratio of classes to discount the majority samples. This strategy is adapted to the popular feature extraction methods such as PCA and NMF. The main advantage of the proposed methods is that they are able to lessen the inherent bias of the extracted features to the majority class in existing PCA and NMF algorithms. Experiments on twelve public datasets with different levels of imbalance ratios show that the proposed methods outperformed the state-of-the-art methods on multiple classifiers.
Castro, EL, Ochoa, EA, Merigo Lindahl, JM & Lafuente, AMG 1970, 'Heavy Moving Averages in exchange rate forecasting', 2016 Annual Conference of the North American Fuzzy Information Processing Society (NAFIPS), 2016 Annual Conference of the North American Fuzzy Information Processing Society (NAFIPS), IEEE, Univ Texas El Paso, El Paso, TX.
View/Download from: Publisher's site
Chalapathy, R, Zare Borzeshi, E & Piccardi, M 1970, 'An Investigation of Recurrent Neural Architectures for Drug Name Recognition', Proceedings of the Seventh International Workshop on Health Text Mining and Information Analysis, Proceedings of the Seventh International Workshop on Health Text Mining and Information Analysis, Association for Computational Linguistics, Austin, Texas, USA.
View/Download from: Publisher's site
Chalapathy, R, Zare Borzeshi, E & Piccardi, M 1970, 'Bidirectional LSTM-CRF for Clinical Concept Extraction', Proceedings for the Clinical NLP workshop - Bidirectional LSTM-CRF for Clinical Concept Extraction, Clinical Natural Language Programming Workshop, ClinicalNLP, Osaka, Japan.
Chang, X, Yang, Y, Long, G, Zhang, C & Hauptmann, AG 1970, 'Dynamic Concept Composition for Zero-Example Event Detection', 30th AAAI Conference on Artificial Intelligence, AAAI 2016, AAAI Conference on Artificial Intelligence, Association for the Advancement of Artificial Intelligence, Phoenix, Arizona, United States, pp. 3464-3470.
View description>>
In this paper, we focus on automatically detecting events in unconstrainedvideos without the use of any visual training exemplars. In principle,zero-shot learning makes it possible to train an event detection model based onthe assumption that events (e.g. \emph{birthday party}) can be described bymultiple mid-level semantic concepts (e.g. 'blowing candle', 'birthday cake').Towards this goal, we first pre-train a bundle of concept classifiers usingdata from other sources. Then we evaluate the semantic correlation of eachconcept \wrt the event of interest and pick up the relevant conceptclassifiers, which are applied on all test videos to get multiple predictionscore vectors. While most existing systems combine the predictions of theconcept classifiers with fixed weights, we propose to learn the optimal weightsof the concept classifiers for each testing video by exploring a set of onlineavailable videos with free-form text descriptions of their content. To validatethe effectiveness of the proposed approach, we have conducted extensiveexperiments on the latest TRECVID MEDTest 2014, MEDTest 2013 and CCV dataset.The experimental results confirm the superiority of the proposed approach.
Chen, Q, Lan, C, Li, J, Chen, B, Wang, L & Zhang, C 1970, 'Depth-First Search Encoding of RNA Substructures', Intelligent Computing Theories and Application, International Conference on Intelligent Computing (ICIC), Springer International Publishing, Lanzhou, China, pp. 328-334.
View/Download from: Publisher's site
View description>>
© Springer International Publishing Switzerland 2016.RNA structural motifs are important in RNA folding process. Traditional index-based and shape-based schemas are useful in modeling RNA secondary structures but ignore the structural discrepancy of individual RNA family member. Further, the in-depth analysis of underlying substructure pattern is underdeveloped owing to varied and unnormalized substructures. This prevents us from understanding RNAs functions. This article proposes a DFS (depth-first search) encoding for RNA substructures. The results show that our methods are useful in modelling complex RNA secondary structures.
Chen, Y, Li, X, Li, L, Liu, G & Xu, G 1970, 'Modeling User Mobility via User Psychological and Geographical Behaviors Towards Point of-Interest Recommendation', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Database Systems for Advanced Applications, Springer International Publishing, Dallas, Texas, USA, pp. 364-380.
View/Download from: Publisher's site
View description>>
© Springer International Publishing Switzerland 2016. The pervasive employments of Location-based Social Network call for precise and personalized Point-of-Interest (POI) recommendation to predict which places the users prefer. Modeling user mobility, as an important component of understanding user preference, plays an essential role in POI recommendation. However, existing methods mainly model user mobility through analyzing the check-in data and formulating a distribution without considering why a user checks in at a specific place from psychological perspective. In this paper, we propose a POI recommendation algorithm modeling user mobility by considering check-in data and geographical information. Specifically, with check-in data, we propose a novel probabilistic latent factor model to formulate user psychological behavior from the perspective of utility theory, which could help reveal the inner information underlying the comparative choice behaviors of users. Geographical behavior of all the historical check-ins captured by a power law distribution is then combined with probabilistic latent factor model to form the POI recommendation algorithm. Extensive evaluation experiments conducted on two real-world datasets confirm the superiority of our approach over state-of-the-art methods.
Chinchore, A, Xu, G & Jiang, F 1970, 'Classifying Sybil in MSNs using C4.5', 2016 INTERNATIONAL CONFERENCE ON BEHAVIORAL, ECONOMIC AND SOCIO-CULTURAL COMPUTING (BESC), IEEE/ACM International Conference on Behavioral, Economic, Socio-Cultural Computing (BESC), IEEE, Durham, NC, pp. 145-150.
Cho, N-H, Wu, Q, Xu, J & Zhang, J 1970, 'Content Authoring Using Single Image in Urban Environments for Augmented Reality', 2016 International Conference on Digital Image Computing: Techniques and Applications (DICTA), 2016 International Conference on Digital Image Computing: Techniques and Applications (DICTA), IEEE, Gold Coast, Australia, pp. 1-7.
View/Download from: Publisher's site
View description>>
© 2016 IEEE. Content authoring is one of essentials of Augmented Reality (AR), which is to emplace an augmented content on a true part of a real scene in order to enhance users' visual experience. For the case of street view single 2D images, the challenge emerges because of clutter environments and unknown position and orientation related to camera pose. Although existing methods based on 2D feature point matching or vanishing point registration may recover the camera pose, the robustness is always challenging because of the uncertainty of feature point detection on texture-less region and displacement of vanishing point detection caused by irregular lines detected on the scene. By taking the advantages of characteristics of the man-made object (e.g. building) widely seen on the street view, this paper proposes a simple yet efficient content authoring approach. In this approach, the building dominant plane where the virtual object will be emplaced is detected and then projected to the frontal-parallel view on which the virtual object can be reliably emplaced. Once the virtual object and the true scene are embedded to each other on the frontal-parallel view, they are able to be converted back to the original view using inverse projection without any distortion. Experiments on public databases show that the proposed method can recover camera pose and implement content emplacement with promising performance.
Dileep Kumar, K, Krishna Reddy, P, Balakrishna Reddy, P & Cao, L 1970, 'Improving the Performance of Collaborative Filtering with Category-Specific Neighborhood', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Asian Conference on Intelligent Information and Database Systems (ACIIDS), Springer Berlin Heidelberg, Da Nang, Vietnam, pp. 201-210.
View/Download from: Publisher's site
View description>>
© Springer-Verlag Berlin Heidelberg 2016.Recommender system (RS) helps customers to select appropriate products from millions of products and has become a key component in e-commerce systems. Collaborative filtering (CF) based approaches are widely employed to build RSs. In CF, recommendation to the target user is computed after forming the corresponding neighbourhood of users. Neighborhood of a target user is extracted based on the similarity between the product rating vector of the target user and the product rating vectors of individual users. In CF, the methodology employed for neighborhood formation influences the performance. In this paper, we have made an effort to improve the performance of CF by proposing a different approach to compute recommendations by considering two kinds of neighborhood. One is the neighborhood by considering the product ratings of the user as a single vector and the other is based on the neighborhood of the corresponding virtual users. For the target user, the virtual users are formed by dividing the ratings based on the category of products. We have proposed a combined approach to compute better recommendations by considering both kinds of neighborhoods. The experiments results on real world MovieLens dataset show that the proposed approach improves the performance over CF.
Fan, X, Xu, RYD & Cao, L 1970, 'Copula mixed-membership stochastic block model', IJCAI International Joint Conference on Artificial Intelligence, International Joint Conference on Artificial Intelligence, AAAI Press / International Joint Conferences on Artificial Intelligence, New York City, New York, United States, pp. 1462-1468.
View description>>
The Mixed-Membership Stochastic Blockmodels (MMSB) is a popular framework for modelling social relationships by fully exploiting each individual node's participation (or membership) in a social network. Despite its powerful representations, MMSB assumes that the membership indicators of each pair of nodes (i.e., people) are distributed independently. However, such an assumption often does not hold in real-life social networks, in which certain known groups of people may correlate with each other in terms of factors such as their membership categories. To expand MMSB's ability to model such dependent relationships, a new framework - a Copula Mixed-Membership Stochastic Blockmodel - is introduced in this paper for modeling intra-group correlations, namely an individual Copula function jointly models the membership pairs of those nodes within the group of interest. This framework enables various Copula functions to be used on demand, while maintaining the membership indicator's marginal distribution needed for modelling membership indicators with other nodes outside of the group of interest. Sampling algorithms for both the finite and infinite number of groups are also detailed. Our experimental results show its superior performance in capturing group interactions when compared with the baseline models on both synthetic and real world datasets.
Fang, M, Yin, J, Zhu, X & Zhang, C 1970, 'TrGraph: Cross-network transfer learning via common signature subgraphs', 2016 IEEE 32nd International Conference on Data Engineering (ICDE), 2016 IEEE 32nd International Conference on Data Engineering (ICDE), IEEE, Helsinki, Finland, pp. 1534-1535.
View/Download from: Publisher's site
View description>>
In this paper, we present a novel transfer learning framework for network node classification. Our objective is to accurately predict node labels in a target network by leveraging information from an auxiliary source network. Such a transfer learning framework is potentially useful for broader areas of network classification, where emerging new networks might not have sufficient labeled information because node labels are either costly to obtain or simply not available, whereas many established networks from related domains are available to benefit the learning. In reality, the source and the target networks may not share common nodes or connections, so the major challenge of cross-network transfer learning is to identify knowledge/patterns transferable between networks and potentially useful to support cross-network learning. In this work, we propose to learn common signature subgraphs between networks, and use them as structure features for the target network. By combining the original node content features and the new structure features, we develop an iterative classification algorithm, TrGraph, that utilizes label dependency to jointly classify nodes in the target network. Experiments on real-world networks demonstrate that TrGraph achieves the superior performance compared to the state-of-the-art baseline methods.
Gao, F & Musial-Gabrys, K 1970, 'Hybrid structure-based link prediction model', 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM), 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM), IEEE, San Francisco, CA, pp. 1221-1228.
View/Download from: Publisher's site
Ghosh, S, Nguyen, H & Li, J 1970, 'Predicting short-term ICU outcomes using a sequential contrast motif based classification framework', 2016 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), 2016 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), IEEE, Orlando, Florida, USA, pp. 5612-5615.
View/Download from: Publisher's site
View description>>
© 2016 IEEE.Critical ICU events like acute hypotension and septic shock are dangerous complications, leading to multiple organ failures and eventual death. Previously, pattern mining algorithms have been employed for extracting interesting rules in various clinical domains. However, the extracted rules are directly investigated by clinicians for diagnosing a disease. Towards this purpose, there is a need to develop advanced prediction models which integrate dynamic patterns to learn a patient's physiological condition. In this study, a sequential contrast patterns-based classification framework is presented for detecting critical patient events, like hypotension and septic shock. Initially, a set of sequential patterns are obtained by using a contrast mining algorithm. Later, these patterns undergo post-processing, for conversion to two novel representations-(1) frequency-based feature space and (2) ordered sequences of patterns, which conserve positional information of a pattern in a time series sequence. Each of these representations are automatically used for developing classification models using SVM and HMM methods. Our results on hypotension and septic shock datasets from a large scale ICU database demonstrate better predictive capabilities, when sequential patterns are used as features.
Han, B, Tsang, IW & Chen, L 1970, 'On the Convergence of a Family of Robust Losses for Stochastic Gradient Descent', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), The European Conference on Machine Learning and Principles and Practice of Knowledge Discovery (ECML-PKDD), Springer International Publishing, Riva del Garda, Italy, pp. 665-680.
View/Download from: Publisher's site
View description>>
© Springer International Publishing AG 2016. The convergence of Stochastic Gradient Descent (SGD) using convex loss functions has been widely studied. However, vanilla SGD methods using convex losses cannot perform well with noisy labels, which adversely affect the update of the primal variable in SGD methods. Unfortunately, noisy labels are ubiquitous in real world applications such as crowdsourcing. To handle noisy labels, in this paper, we present a family of robust losses for SGD methods. By employing our robust losses, SGD methods successfully reduce negative effects caused by noisy labels on each update of the primal variable. We not only reveal the convergence rate of SGD methods using robust losses, but also provide the robustness analysis on two representative robust losses. Comprehensive experimental results on six real-world datasets show that SGD methods using robust losses are obviously more robust than other baseline methods in most situations with fast convergence.
Hazber, MAG, Li, R, Xu, G & Alalayah, KM 1970, 'An Approach for Automatically Generating R2RML-Based Direct Mapping from Relational Databases', Communications in Computer and Information Science, International Conference of Young Computer Scientists, Engineers and Educators (ICYCSEE), Springer Singapore, Harbin, China, pp. 151-169.
View/Download from: Publisher's site
View description>>
For integrating relational databases (RDBs) into semantic web applications, the W3C RDB2RDF Working Group recommended two approaches, Direct Mapping (DM) and R2RML. The DM provides a set of mapping rules according to RDB schema, while the R2RML allows users to manually define mappings according to existing target ontology. The major problem to use R2RML is the effort for creating R2RML mapping documents manually. This may lead to appearance of many mistakes in the R2RML documents and requires domain experts. In this paper, we propose and implement an approach to generate an R2RML mapping documents automatically from RDB schema. The R2RML mapping reflects the behavior of the DM specification and allows any R2RML parser to generate a set of RDF triples from relational data. The input of generating approach is DBsInfo class that automatically generated from relational schema. An experimental prototype is developed and shows the effectiveness of our approach algorithms.
He, H, Maple, C, Watson, T, Tiwari, A, Mehnen, J, Jin, Y & Gabrys, B 1970, 'The security challenges in the IoT enabled cyber-physical systems and opportunities for evolutionary computing & other computational intelligence', 2016 IEEE Congress on Evolutionary Computation (CEC), 2016 IEEE Congress on Evolutionary Computation (CEC), IEEE, Vancouver, CANADA, pp. 1015-1021.
View/Download from: Publisher's site
View description>>
Internet of Things (IoT) has given rise to the fourth industrial revolution (Industrie 4.0), and it brings great benefits by connecting people, processes and data. However, cybersecurity has become a critical challenge in the IoT enabled cyber physical systems, from connected supply chain, Big Data produced by huge amount of IoT devices, to industry control systems. Evolutionary computation combining with other computational intelligence will play an important role for cybersecurity, such as artificial immune mechanism for IoT security architecture, data mining/fusion in IoT enabled cyber physical systems, and data driven cybersecurity. This paper provides an overview of security challenges in IoT enabled cyber-physical systems and what evolutionary computation and other computational intelligence technology could contribute for the challenges. The overview could provide clues and guidance for research in IoT security with computational intelligence.
He, H, Tiwari, A, Mehnen, J, Watson, T, Maple, C, Jin, Y & Gabrys, B 1970, 'Incremental information gain analysis of input attribute impact on RBF-kernel SVM spam detection', 2016 IEEE Congress on Evolutionary Computation (CEC), 2016 IEEE Congress on Evolutionary Computation (CEC), IEEE, Vancouver, CANADA, pp. 1022-1029.
View/Download from: Publisher's site
View description>>
The massive increase of spam is posing a very serious threat to email and SMS, which have become an important means of communication. Not only do spams annoy users, but they also become a security threat. Machine learning techniques have been widely used for spam detection. Email spams can be detected through detecting senders' behaviour, the contents of an email, subject and source address, etc, while SMS spam detection usually is based on the tokens or features of messages due to short content. However, a comprehensive analysis of email/SMS content may provide cures for users to aware of email/SMS spams. We cannot completely depend on automatic tools to identify all spams. In this paper, we propose an analysis approach based on information entropy and incremental learning to see how various features affect the performance of an RBF-based SVM spam detector, so that to increase our awareness of a spam by sensing the features of a spam. The experiments were carried out on the spambase and SMSSpemCollection databases in UCI machine learning repository. The results show that some features have significant impacts on spam detection, of which users should be aware, and there exists a feature space that achieves Pareto efficiency in True Positive Rate and True Negative Rate.
Huang, X, Fan, L, Zhang, J, Wu, Q & Yuan, C 1970, 'Real Time Complete Dense Depth Reconstruction for a Monocular Camera', 2016 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), 2016 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), IEEE, Las Vegas, Nevada., pp. 674-679.
View/Download from: Publisher's site
View description>>
© 2016 IEEE. In this paper, we aim to solve the problem of estimating complete dense depth maps from a monocular moving camera. By 'complete', we mean depth information is estimated for every pixel and detailed reconstruction is achieved. Although this problem has previously been attempted, the accuracy of complete dense depth reconstruction is a remaining problem. We propose a novel system which produces accurate complete dense depth map. The new system consists of two subsystems running in separated threads, namely, dense mapping and sparse patch-based tracking. For dense mapping, a new projection error computation method is proposed to enhance the gradient component in estimated depth maps. For tracking, a new sparse patch-based tracking method estimates camera pose by minimizing a normalized error term. The experiments demonstrate that the proposed method obtains improved performance in terms of completeness and accuracy compared to three state-of the-art dense reconstruction methods VSFM+CMVC, LSDSLAM and REMODE.
Huang, X, Zhang, J, Wu, Q, Fan, L & Yuan, C 1970, 'A coarse-to-fine algorithm for registration in 3D street-view cross-source point clouds', 2016 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2016, Digital Image Computing Techniques and Applications, IEEE, Gold coast, Australia..
View/Download from: Publisher's site
View description>>
With the development of numerous 3D sensing technologies, object registrationon cross-source point cloud has aroused researchers' interests. When the pointclouds are captured from different kinds of sensors, there are large anddifferent kinds of variations. In this study, we address an even morechallenging case in which the differently-source point clouds are acquired froma real street view. One is produced directly by the LiDAR system and the otheris generated by using VSFM software on image sequence captured from RGBcameras. When it confronts to large scale point clouds, previous methods mostlyfocus on point-to-point level registration, and the methods have manylimitations.The reason is that the least mean error strategy shows poor abilityin registering large variable cross-source point clouds. In this paper,different from previous ICP-based methods, and from a statistic view, wepropose a effective coarse-to-fine algorithm to detect and register a smallscale SFM point cloud in a large scale Lidar point cloud. Seen from theexperimental results, the model can successfully run on LiDAR and SFM pointclouds, hence it can make a contribution to many applications, such as roboticsand smart city development.
Hussein, F, Awwad, S, Piccardi, M & IEEE 1970, 'JOINT ACTION RECOGNITION AND SUMMARIZATION BY SUB-MODULAR INFERENCE', 2016 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING PROCEEDINGS, IEEE International Conference on Acoustics, Speech and Signal Processing, IEEE, Shanghai, China, pp. 2697-2701.
View/Download from: Publisher's site
View description>>
Action recognition and video summarization are two important multimedia tasks that are useful for applications such as video indexing and retrieval, video surveillance, human-computer interaction and home intelligence. While many approaches exist in the literature for these two tasks, to date they have always been addressed separately. Instead, in this paper we move from the assumption that these two tasks should be tackled as a joint objective: on the one hand, action recognition can drive the selection of meaningful and informative summaries; on the other, recognizing actions from a summary rather than the entire video can in principle reduce noise and prove more accurate. To this aim, we propose a novel approach for joint action recognition-summarization based on the performing latent structural SVM framework, together with an efficient algorithm for inferring the action and the summary based on the property of sub-modularity. Experimental results on a challenging benchmark, MSR Dai-lyActivity3D, show that the approach is capable of achieving remarkable action recognition accuracy while providing appealing video summaries.
Jiang, F, Gan, J, Xu, Y & Xu, G 1970, 'Coupled behavioral analysis for user preference-based email spamming', 2016 International Conference on Behavioral, Economic and Socio-cultural Computing (BESC), 2016 International Conference on Behavioral, Economic and Socio-cultural Computing (BESC), IEEE, Durham, NC, pp. 72-76.
View/Download from: Publisher's site
View description>>
© 2016 IEEE.In this paper, we develop and implement a new email spamming system leveraged by coupled text similarity analysis on user preference and a virtual meta-layer user-based email network, we take the social networks or campus LAN networks as the spam social network scenario. Fewer current practices exploit social networking initiatives to assist in spam filtering. Social network has essentially a large number of accounts features and attributes to be considered. Instead of considering large amount of users accounts features, we construct a new model called meta-layer email network which can reduce these features by only considering individual user's actions as an indicator of user preference, these common user actions are considered to construct a social behavior-based email network. With the further analytic results from text similarity measurements for each individual email contents, the behavior-based virtual email network can be improved with much higher accuracy on user preferences. Further, a coupled selection model is developed for this email network, we are able to consider all relevant factors/features in a whole and recommend the emails practically to the user individually. The experimental results show the new approach can achieve higher precision and accuracy with better email ranking in favor of personalised preference.
Kuppili Venkata, S, Keppens, J & Musial, K 1970, 'Agent Based Simulation to Evaluate Adaptive Caching in Distributed Databases', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Springer International Publishing, pp. 455-462.
View/Download from: Publisher's site
View description>>
Caching frequently used data is a common practice to improve query performance in database systems. But traditional algorithms used for cache management prove to be insufficient in distributed environment where groups of users require similar or related data from multiple databases. Repeated data transfers can become a bottleneck leading to long query response time and high resource utilization. Our work focuses on adaptive algorithms to decide on optimal grain of data to be cached and cache refreshment techniques to reduce data transfers. In this paper, we present agent based simulation to investigate and in consequence improve cache management in the distributed database environment. Dynamic grain size and decisions on cache refreshment are made as a result of coordination and interaction between agents. Initial results show better response time and higher data availability compared to traditional caching techniques.
Liu, C & Chen, L 1970, 'Summarizing uncertain transaction databases by Probabilistic Tiles', 2016 International Joint Conference on Neural Networks (IJCNN), 2016 International Joint Conference on Neural Networks (IJCNN), IEEE, Vancouver, pp. 4375-4382.
View/Download from: Publisher's site
View description>>
© 2016 IEEE. Transaction data mining is ubiquitous in various domains and has been researched extensively. In recent years, observing that uncertainty is inherent in many real world applications, uncertain data mining has attracted much research attention. Among the research problems, summarization is important because it produces concise and informative results, which facilitates further analysis. However, there are few works exploring how to effectively summarize uncertain transaction data. In this paper, we formulate the problem of summarizing uncertain transaction data as Minimal Probabilistic Tile Cover Mining, which aims to find a high-quality probabilistic tile set covering an uncertain database with minimal cost. We define the concept of Probabilistic Price and Probabilistic Price Order to evaluate and compare the quality of tiles, and propose a framework to discover the minimal probabilistic tile cover. The bottleneck is to check whether a tile is better than another according to the Probabilistic Price Order, which involves the computation of a joint probability. We prove that it can be decomposed into independent terms and calculated efficiently. Several optimization techniques are devised to further improve the performance. Experimental results on real world datasets demonstrate the conciseness of the produced tiles and the effectiveness and efficiency of our approach.
Liu, W, Chen, X, Yang, J & Wu, Q 1970, 'Robust weighted least squares for guided depth upsampling', 2016 IEEE International Conference on Image Processing (ICIP), 2016 IEEE International Conference on Image Processing (ICIP), IEEE, Phoenix, AZ, USA, pp. 559-563.
View/Download from: Publisher's site
View description>>
© 2016 IEEE. In this paper, we propose a new guided depth upsampling method denoted as Robust Weighted Least Squares (RWLS). Our work is inspired by the connection between the Weighted Least Squares (WLS) and the Auto Regressive (AR) model. By adopting a new robust penalty function to model the smoothness of the proposed model, we show that the proposed method performs much better in preserving sharp depth discontinuities than previous work. Through both mathematical analysis and experimental results, we show that our method has promising performance on handling the inconsistency between the guidance image and the depth map in both preserving sharp depth discontinuities and suppressing the texture copy artifacts.
Manongdo, R & Xu, G 1970, 'Applying client churn prediction modeling on home-based care services industry', 2016 International Conference on Behavioral, Economic and Socio-cultural Computing (BESC), 2016 International Conference on Behavioral, Economic and Socio-cultural Computing (BESC), IEEE, Durham, NC, pp. 167-172.
View/Download from: Publisher's site
View description>>
Client churn prediction model is widely acknowledged as an effective way of realizing customer life-time value especially in service-oriented industries and under a competitive business environment. Churn model allows targeting of clients for retention campaigns and is a critical component of customer relationship management(CRM) and business intelligence systems. There are numerous statistical models and techniques applied successfully on data mining projects for various industries. While there is literature for prediction modeling on hospital health care services, non-exist for home-based care services. In this study, logistic regression, random forest and C5.0 decision tree were the models used in building a binary client churn classifier for a home-based care services company based in Australia. All models yielded prediction accuracies over 90% with tree based classifiers marginally higher and C5.0 model found to be suitable for use in this industry. This study also showed that existing client satisfaction measures currently in use by the company does not adequately contribute to churn analysis.
Martin Salvador, M, Budka, M & Gabrys, B 1970, 'Towards Automatic Composition of Multicomponent Predictive Systems', Hybrid Artificial Intelligent Systems, International Conference on Hybrid Artificial Intelligence Systems, Springer International Publishing, Seville, Spain, pp. 27-39.
View/Download from: Publisher's site
View description>>
Automatic composition and parametrisation of multicomponent predictive systems (MCPSs) consisting of chains of data transformation steps is a challenging task. In this paper we propose and describe an extension to the Auto-WEKA software which now allows to compose and optimise such flexible MCPSs by using a sequence of WEKA methods. In the experimental analysis we focus on examining the impact of significantly extending the search space by incorporating additional hyperparameters of the models, on the quality of the found solutions. In a range of extensive experiments three different optimisation strategies are used to automatically compose MCPSs on 21 publicly available datasets. A comparison with previous work indicates that extending the search space improves the classification accuracy in the majority of the cases. The diversity of the found MCPSs are also an indication that fully and automatically exploiting different combinations of data cleaning and preprocessing techniques is possible and highly beneficial for different predictive models. This can have a big impact on high quality predictive models development, maintenance and scalability aspects needed in modern application and deployment scenarios.
Merigo, JM, Alrajeh, N & Peris-Ortiz, M 1970, 'Induced aggregation operators in the ordered weighted average sum', 2016 IEEE Symposium Series on Computational Intelligence (SSCI), 2016 IEEE Symposium Series on Computational Intelligence (SSCI), IEEE, Athens, Greece, pp. 1-6.
View/Download from: Publisher's site
View description>>
© 2016 IEEE. The ordered weighted average (OWA) aggregation is an extension of the classical weighted average by using a reordering process of the arguments in a decreasing or increasing way. This article presents new averaging aggregation operators by using sums and order inducing variables. This approach produces the induced ordered weighted average sum (IOWAS). The IOWAS operator aggregates a set of sums using a complex reordering process based on order-inducing variables. This approach includes a different types of aggregation structures including the well-known OWA families. The work presents additional generalizations by using generalized and quasi-arithmetic means. The paper ends with a simple numerical example that shows how to aggregate with this new approach.
Merigo, JM, Blanco-Mesa, F, Gil-Lafuente, AM & Yager, RR 1970, 'A bibliometric analysis of the first thirty years of the International Journal of Intelligent Systems', 2016 IEEE Symposium Series on Computational Intelligence (SSCI), 2016 IEEE Symposium Series on Computational Intelligence (SSCI), IEEE, Athens, Greece.
View/Download from: Publisher's site
View description>>
© 2016 IEEE. The International Journal of Intelligent Systems was created on 1986. Today, the journal has become thirty years old. In order to celebrate this anniversary, this study develops a bibliometric review of all the papers published in the journal between 1986 and 2015. The results are mainly based on the Web of Science Core Collection that classifies the bibliographic material by using several indicators including the total number of publications and citations, the h-index, the cites per paper and the citing articles. Moreover, the work also uses the VOS viewer software for visualizing the main results through bibliographic coupling and co-citation. The results show a general overview of the leading trends that have influenced the journal in terms of highly cited papers, authors, journals, universities and countries.
Merigó, JM, Zurita, G & Link-Chaparro, S 1970, 'Normalization of the article influence score between categories', Lecture Notes in Engineering and Computer Science, pp. 182-187.
View description>>
This study introduces a normalized article influence score. The main objective is to show that the article influence score obtained in different categories is not equivalent and it is necessary to normalize it when comparing journals form different categories. Several methods are suggested including a normalization that divides the article influence score by the average and another approach that normalizes the results in [0, 1] inside the same category in order to be able to compare between different fields. The results show that each category have different results and it is necessary to develop a normalization process in order to compare the journals. The article analyses a case study in engineering.
Merigó, JM, Zurita, G & Lobos-Ossandón, V 1970, 'Computer science research in artificial intelligence', Lecture Notes in Engineering and Computer Science, pp. 216-220.
View description>>
This paper presents a bibliometric overview of the research carried out between 1990 and 2014 in computer science with a focus on artificial intelligence. The work analyses all the journals available in Web of Science during this period and presents their publication and citation results. The study also considers the most cited articles in this area during the last twenty-five years. IEEE Journals obtain the most remarkable results publishing more than half of the most cited papers.
Montgomery, J, Reid, M & Drake, BJ 1970, 'Protocols and Structures for Inference: A RESTful API for Machine Learning', Proceedings of The 2nd International Conference on Predictive APIs and Apps, 2nd International Conference on Predictive APIs and Apps, Journal of Machine Learning Research, Sydney, pp. 29-42.
View description>>
Diversity in machine learning APIs (in both software toolkits and web services), works against realising machine learning’s full potential, making it difficult to draw on individual algorithms from different products or to compose multiple algorithms to solve complex tasks. This paper introduces the Protocols and Structures for Inference (PSI) service architecture and specification, which presents inferential entities—relations, attributes, learners and predictors—as RESTful web resources that are accessible via a common but flexible and extensible interface. Resources describe the data they ingest or emit using a variant of the JSON schema language, and the API has mechanisms to support non-JSON data and future extension of service features.
Ochoa, EA, Castro, EL, Lindahl, JMM & Lafuente, AMG 1970, 'Forgotten effects and heavy moving averages in exchange rate forecasting', 2016 IEEE Symposium Series on Computational Intelligence (SSCI), 2016 IEEE Symposium Series on Computational Intelligence (SSCI), IEEE, Athens, Greece, pp. 1-7.
View/Download from: Publisher's site
View description>>
© 2016 IEEE. This paper presents the results of using experton, forgotten effects and heavy moving averages operators in three traditional models based purchasing power parity (PPP) model to forecast exchange rate. Therefore, the use of these methods is to improve the forecast error under scenarios of volatility and uncertainty, such as the financial markets and more precise in exchange rate. The heavy ordered weighted moving average weighted average (HOWMAWA) operator is introduced. This new operator includes the weighted average in the usual heavy ordered weighted moving average (HOWMA) operator, considering a degree of importance for each concept that includes the operator. The use of experton and forgotten effects methodology represents the information of the experts in the field and with that information were obtained hidden variables or second degree relations. The results show that the inclusion of the forgotten effects and heavy moving average operators improve our results and reduce the forecast error.
Pan, S, Wu, J, Zhu, X, Zhang, C & Wang, Y 1970, 'Tri-party deep network representation', IJCAI International Joint Conference on Artificial Intelligence, International Joint Conference on Artificial Intelligence, AAAI Press / International Joint Conferences on Artificial Intelligence, New York City, New York, United States, pp. 1895-1901.
View description>>
Information network mining often requires examination of linkage relationships between nodes for analysis. Recently, network representation has emerged to represent each node in a vector format, embedding network structure, so off-the-shelf machine learning methods can be directly applied for analysis. To date, existing methods only focus on one aspect of node information and cannot leverage node labels. In this paper, we propose TriDNR, a tri-party deep network representation model, using information from three parties: node structure, node content, and node labels (if available) to jointly learn optimal node representation. TriDNR is based on our new coupled deep natural language module, whose learning is enforced at three levels: (1) at the network structure level, TriDNR exploits inter-node relationship by maximizing the probability of observing surrounding nodes given a node in random walks; (2) at the node content level, TriDNR captures node-word correlation by maximizing the co-occurrence of word sequence given a node; and (3) at the node label level, TriDNR models label-word correspondence by maximizing the probability of word sequence given a class label. The tri-party information is jointly fed into the neural network model to mutually enhance each other to learn optimal representation, and results in up to 79% classification accuracy gain, compared to state-of-the-art methods.
Pan, S, Wu, J, Zhuy, X, Zhang, C & Yuz, PS 1970, 'Joint structure feature exploration and regularization for multi-task graph classification', 2016 IEEE 32nd International Conference on Data Engineering (ICDE), 2016 IEEE 32nd International Conference on Data Engineering (ICDE), IEEE, Helsinki, Finland, pp. 1474-1475.
View/Download from: Publisher's site
View description>>
We formulate a new multi-task graph classification
(MTG) problem, where multiple graph classification tasks are
jointly regularized to find discriminative subgraphs shared by all
tasks for learning. More details can be found in [1].
Pang, G, Cao, L & Chen, L 1970, 'Outlier detection in complex categorical data by modelling the feature value couplings', IJCAI International Joint Conference on Artificial Intelligence, International Joint Conference on Artificial Intelligence (IJCAI), AAAI Press, New York, pp. 1902-1908.
View description>>
This paper introduces a novel unsupervised outlier detection method, namely Coupled Biased Random Walks (CBRW), for identifying outliers in categorical data with diversified frequency distributions and many noisy features. Existing pattern-based outlier detection methods are ineffective in handling such complex scenarios, as they misfit such data. CBRW estimates outlier scores of feature values by modelling feature value level couplings, which carry intrinsic data characteristics, via biased random walks to handle this complex data. The outlier scores of feature values can either measure the outlierness of an object or facilitate the existing methods as a feature weighting and selection indicator. Substantial experiments show that CBRW can not only detect outliers in complex data significantly better than the state-of-the-art methods, but also greatly improve the performance of existing methods on data sets with many noisy features.
Pang, G, Cao, L, Chen, L & Liu, H 1970, 'Unsupervised Feature Selection for Outlier Detection by Modelling Hierarchical Value-Feature Couplings', 2016 IEEE 16th International Conference on Data Mining (ICDM), 2016 IEEE 16th International Conference on Data Mining (ICDM), IEEE, Barcelona, pp. 410-419.
View/Download from: Publisher's site
View description>>
© 2016 IEEE. Proper feature selection for unsupervised outlier detection can improve detection performance but is very challenging due to complex feature interactions, the mixture of relevant features with noisy/redundant features in imbalanced data, and the unavailability of class labels. Little work has been done on this challenge. This paper proposes a novel Coupled Unsupervised Feature Selection framework (CUFS for short) to filter out noisy or redundant features for subsequent outlier detection in categorical data. CUFS quantifies the outlierness (or relevance) of features by learning and integrating both the feature value couplings and feature couplings. Such value-To-feature couplings capture intrinsic data characteristics and distinguish relevant features from those noisy/redundant features. CUFS is further instantiated into a parameter-free Dense Subgraph-based Feature Selection method, called DSFS. We prove that DSFS retains a 2-Approximation feature subset to the optimal subset. Extensive evaluation results on 15 real-world data sets show that DSFS obtains an average 48% feature reduction rate, and enables three different types of pattern-based outlier detection methods to achieve substantially better AUC improvements and/or perform orders of magnitude faster than on the original feature set. Compared to its feature selection contender, on average, all three DSFS-based detectors achieve more than 20% AUC improvement.
Peng, F, Lu, X, Lu, J, Xu, RY-D, Luo, C, Ma, C & Yang, J 1970, 'MetricRec: Metric Learning for Cold-Start Recommendations', Advanced Data Mining and Applications (LNAI), International Conference on Advanced Data Mining and Applications, Springer International Publishing, Gold Coast, QLD, Australia, pp. 445-458.
View/Download from: Publisher's site
View description>>
© Springer International Publishing AG 2016.Making recommendations for new users is a challenging task of cold-start recommendations due to the absence of historical ratings. When the attributes of users are available, such as age, occupation and gender, then new users’ preference can be inferred. Inspired by the user based collaborative filtering in warm-start scenario, we propose using the similarity on attributes to conduct recommendations for new users. Two basic similarity metrics, cosine and Jaccard, are evaluated for cold-start. We also propose a novel recommendation model, MetricRec, that learns an interest-derived metric such that the users with similar interests are close to each other in the attribute space. As the MetricRec’s feasible area is conic, we propose an efficient Interior-point Stochastic Gradient Descent (ISGD) method to optimize it. During the optimizing process, the metric is always guaranteed in the feasible area. Owing to the stochastic strategy, ISGD possesses scalability. Finally, the proposed models are assessed on two movie datasets, Movielens-100K and Movielens-1M. Experimental results demonstrate that MetricRec can effectively learn the interest-derived metric that is superior to cosine and Jaccard, and solve the cold-start problem effectively.
Poostchi, H, Borzeshi, EZ, Abdous, M & Piccardi, M 1970, 'PersoNER: Persian named-entity recognition', COLING 2016 - 26th International Conference on Computational Linguistics, Proceedings of COLING 2016: Technical Papers, International Conference on Computational Linguistics, COLING, Osaka, Japan, pp. 3381-3389.
View description>>
Named-Entity Recognition (NER) is still a challenging task for languages with low digital resources. The main difficulties arise from the scarcity of annotated corpora and the consequent problematic training of an effective NER pipeline. To abridge this gap, in this paper we target the Persian language that is spoken by a population of over a hundred million people world-wide. We first present and provide ArmanPerosNERCorpus, the first manually-annotated Persian NER corpus. Then, we introduce PersoNER, an NER pipeline for Persian that leverages a word embedding and a sequential max-margin classifier. The experimental results show that the proposed approach is capable of achieving interesting MUC7 and CoNNL scores while outperforming two alternatives based on a CRF and a recurrent neural network.
Qiao, M, Bian, W, Xu, RYD & Tao, D 1970, 'Diversified hidden Markov models for sequential labeling', 2016 IEEE 32nd International Conference on Data Engineering (ICDE), 2016 IEEE 32nd International Conference on Data Engineering (ICDE), IEEE, Helsinki, FINLAND, pp. 1512-+.
View/Download from: Publisher's site
Qin, Z, Wu, J, Hong, Y, Tian, Y & Chengqi, Z 1970, 'Unsupervised feature learning from time series', IJCAI International Joint Conference on Artificial Intelligence, International Joint Conference on Artificial Intelligence, AAAI Press / International Joint Conferences on Artificial Intelligence, Palo Alto, California, United States, pp. 2322-2328.
View description>>
In this paper we study the problem of learning discriminative features (segments), often referred to as shapelets [Ye and Keogh, 2009] of time series, from unlabeled time series data. Discovering shapelets for time series classification has been widely studied, where many search-based algorithms are proposed to efficiently scan and select segments from a pool of candidates. However, such types of search-based algorithms may incur high time cost when the segment candidate pool is large. Alternatively, a recent work [Grabocka et al., 2014] uses regression learning to directly learn, instead of searching for, shapelets from time series. Motivated by the above observations, we propose a new Unsupervised Shapelet Learning Model (USLM) to efficiently learn shapelets from unlabeled time series data. The corresponding learning function integrates the strengths of pseudo-class label, spectral analysis, shapelets regularization term and regularized least-squares to auto-learn shapelets, pseudo-class labels and classification boundaries simultaneously. A coordinate descent algorithm is used to iteratively solve the learning function. Experiments show that USLM outperforms searchbased algorithms on real-world time series data.
Quyen, NTH, Tung, KT, Hanh, LTM & Binh, NT 1970, 'Improving mutant generation for Simulink models using genetic algorithm', 2016 International Conference on Electronics, Information, and Communications (ICEIC), 2016 International Conference on Electronics, Information, and Communications (ICEIC), IEEE, Danang, VIETNAM.
View/Download from: Publisher's site
Ruiqi Hu, Pan, S, Guodong Long, Xingquan Zhu, Jing Jiang & Chengqi Zhang 1970, 'Co-clustering enterprise social networks', 2016 International Joint Conference on Neural Networks (IJCNN), 2016 International Joint Conference on Neural Networks (IJCNN), IEEE, IEEE, pp. 107-114.
View/Download from: Publisher's site
View description>>
© 2016 IEEE. An enterprise social network (ESN) involves diversified user groups from producers, suppliers, logistics, to end consumers, and users have different scales, broad interests, and various objectives, such as advertising, branding, customer relationship management etc. In addition, such a highly diversified network is also featured with rich content, including recruiting messages, advertisements, news release, customer complains etc. Due to such complex nature, an immediate need is to properly organize a chaotic enterprise social network as functional groups, where each group corresponds to a set of peers with business interactions and common objectives, and further understand the business role of each group, such as their common interests and key features differing from other groups. In this paper, we argue that due to unique characteristics of enterprise social networks, simple clustering for ESN nodes or using existing topic discovery methods cannot effectively discover functional groups and understand their roles. Alternatively, we propose CENFLD, which carries out co-clustering on enterprise social networks for functional group discovery and understanding. CENFLD is a co-factorization based framework which combines network topology structures and rich content information, including interactions between nodes and correlations between node content, to discover functional user groups. Because the number of functional groups is highly data driven and hard to estimate, CENFLD employs a hold-out test principle to find the group number optimally complying with the underlying data. Experiments and comparisons, with state-of-the-art approaches, on 13 real-world enterprise/organizational networks validate the performance of CENFLD.
Shao, J, Meng, X & Cao, L 1970, 'Mining actionable combined high utility incremental and associated patterns', 2016 IEEE International Conference on Aircraft Utility Systems (AUS), 2016 IEEE/CSAA International Conference on Aircraft Utility Systems (AUS), IEEE, Beijing,China, pp. 1164-1169.
View/Download from: Publisher's site
View description>>
High Utility Itemsets(HUI) Mining, instead of Frequent Pattern Mining (FIM), has been an attractive theme in data mining domain for over a decade since it can be regarded as an alternative way for researchers to identify actionable patterns. In addition, the necessity of decision-making actions and behavior-oriented strategies based on large amount of informative data impels the significance of discovering actionable patterns to be widely admitted. The current HUIM research focus has been on improving the efficiency to make algorithms faster and more stable. However, the coupling relationships between items in given itemsets are ignored. For example, the utility of one itemset might be lower than the manager expected until one additional item takes part in; and vice versa, the utility of an itemset might drop sharply when another one joins in. What's more, it is not occasional to find out that quite a lot of redundant itemsets sharing the same underlying item are presented based on existing academic HUI mining methods. Store managers would not make expected profits based on such results which makes the results not actionable at all. To this end, here we introduce a new framework for mining actionable patterns, called Mining Utility Associated Patterns (MUAP), which aims to find high utility incremental and strongly associated item/itemset with combined incorporating criteria. The outputs of this algorithm are convincing on real datasets as well as synthetic datasets.
Song, K, Chen, L, Gao, W, Feng, S, Wang, D & Zhang, C 1970, 'PerSentiment', Proceedings of the 25th International Conference Companion on World Wide Web - WWW '16 Companion, the 25th International Conference Companion, ACM Press, Montreal, pp. 255-258.
View/Download from: Publisher's site
View description>>
Microblogging services are playing increasingly important roles in our daily life today. It is useful for microblog users to instantly understand the sentiment of a large number of microblogs posted by their friends and make appropriate response. Despite considerable progress on microblog sentiment classification, most of the existing works ignore the influence of personal distinctions of different microblog users on the sentiments they convey, and none of them has provided real-world personalized sentiment classification systems. Considering personal distinctions in sentiment analysis is natural and necessary as different people have different language habits, personal characters, opinion bias and so on. In this demonstration, we present a live system based on Twitter called PerSentiment, an individuality-dependent sentiment classification system which makes the first attempt to analyze the personalized sentiment of recent tweets and retweets posted by the authenticated user and the users he/she follows. Our system consists of four steps, i.e., requesting tweets via Twitter API, preprocessing collected tweets for extracting features, building personalized sentiment classifier based on a novel and extensible Latent Factor Model (LFM) trained on emoticon-tagged tweets, and finally visualizing the sentiment of friends' tweets to provide a guide for better sentiment understanding.
Song, K, Gao, W, Chen, L, Feng, S, Wang, D & Zhang, C 1970, 'Build Emotion Lexicon from the Mood of Crowd via Topic-Assisted Joint Non-negative Matrix Factorization', Proceedings of the 39th International ACM SIGIR conference on Research and Development in Information Retrieval, SIGIR '16: The 39th International ACM SIGIR conference on research and development in Information Retrieval, ACM, Pisa, Italy.
View/Download from: Publisher's site
Thuy Do, QN, Zhilin, A, Junior, CZP, Wang, G & Hussain, FK 1970, 'A network-based approach to detect spammer groups', 2016 International Joint Conference on Neural Networks (IJCNN), 2016 International Joint Conference on Neural Networks (IJCNN), IEEE, Vancouver, BC, Canada, pp. 3642-3648.
View/Download from: Publisher's site
View description>>
© 2016 IEEE. Online reviews nowadays are an important source of information for consumers to evaluate online services and products before deciding which product and which provider to choose. Therefore, online reviews have significant power to influence consumers' purchase decisions. Being aware of this, an increasing number of companies have organized spammer review campaigns, in order to promote their products and gain an advantage over their competitors by manipulating and misleading consumers. To make sure the Internet remains a reliable source of information, we propose a method to identify both individual and group spamming reviews by assigning a suspicion score to each user. The proposed method is a network-based approach combining clustering techniques. We demonstrate the efficiency and effectiveness of our approach on a real-world and manipulated dataset that contains over 8000 restaurants and 600,000 restaurant reviews from TripAdvisor website. We tested our method in three testing scenarios. The method was able to detect all spammers in two testing scenarios, however it did not detect all in the last scenario.
Wang, D, Deng, S & Xu, G 1970, 'GEMRec: A Graph-Based Emotion-Aware Music Recommendation Approach', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), International Conference on Web Information Systems Engineering, Springer International Publishing, Shanghai, China, pp. 92-106.
View/Download from: Publisher's site
View description>>
© Springer International Publishing AG 2016. Music recommendation has gained substantial attention in recent times. As one of the most important context features,user emotion has great potential to improve recommendations,but this has not yet been sufficiently explored due to the difficulty of emotion acquisition and incorporation. This paper proposes a graph-based emotion-aware music recommendation approach (GEMRec) by simultaneously taking a user’s music listening history and emotion into consideration. The proposed approach models the relations between user,music,and emotion as a three-element tuple (user,music,emotion),upon which an Emotion Aware Graph (EAG) is built,and then a relevance propagation algorithm based on random walk is devised to rank the relevance of music items for recommendation. Evaluation experiments are conducted based on a real dataset collected from a Chinese microblog service in comparison to baselines. The results show that the emotional context from a user’s microblogs contributes to improving the performance of music recommendation in terms of hitrate,precision,recall,and F1 score.
Wang, D, Deng, S, Liu, S & Xu, G 1970, 'Improving Music Recommendation Using Distributed Representation', Proceedings of the 25th International Conference Companion on World Wide Web - WWW '16 Companion, the 25th International Conference Companion, ACM Press, Montreal, Canada, pp. 125-126.
View/Download from: Publisher's site
View description>>
In this paper, a music recommendation approach based on distributed representation is presented. The proposed approach firstly learns the distributed representations of music pieces and acquires users' preferences from listening records. Then, it recommends appropriate music pieces whose distributed representations are in accordance with target users' preferences. Experiments on a real world dataset demonstrate that the proposed approach outperforms the state-of-the-art methods.
Wang, D, Deng, S, Zhang, X & Xu, G 1970, 'Learning Music Embedding with Metadata for Context Aware Recommendation', Proceedings of the 2016 ACM on International Conference on Multimedia Retrieval, ICMR'16: International Conference on Multimedia Retrieval, ACM, New York, USA, pp. 249-253.
View/Download from: Publisher's site
View description>>
© 2016 ACM. Contextual factors can benefit music recommendation and retrieval tasks remarkably. However, how to acquire and utilize the contextual information still need to be studied. In this paper, we propose a context aware music recommendation approach, which can recommend music appropriate for users' contextual preference for music. In analogy to matrix factorization methods for collaborative filtering, the proposed approach does not require songs to be described by features beforehand, but it learns music pieces' embeddings (vectors in low-dimensional continuous space) from music playing records and corresponding metadata and infer users' general and contextual preference for music from their playing records with the learned embedding. Then, our approach can recommend appropriate music pieces. Experimental evaluations on a real world dataset show that the proposed approach outperforms baseline methods.
Wang, S, Liu, W, Wu, J, Cao, L, Meng, Q & Kennedy, PJ 1970, 'Training deep neural networks on imbalanced data sets', 2016 International Joint Conference on Neural Networks (IJCNN), 2016 International Joint Conference on Neural Networks (IJCNN), IEEE, Vancouver, Canada, pp. 4368-4374.
View/Download from: Publisher's site
View description>>
© 2016 IEEE. Deep learning has become increasingly popular in both academic and industrial areas in the past years. Various domains including pattern recognition, computer vision, and natural language processing have witnessed the great power of deep networks. However, current studies on deep learning mainly focus on data sets with balanced class labels, while its performance on imbalanced data is not well examined. Imbalanced data sets exist widely in real world and they have been providing great challenges for classification tasks. In this paper, we focus on the problem of classification using deep network on imbalanced data sets. Specifically, a novel loss function called mean false error together with its improved version mean squared false error are proposed for the training of deep networks on imbalanced data sets. The proposed method can effectively capture classification errors from both majority class and minority class equally. Experiments and comparisons demonstrate the superiority of the proposed approach compared with conventional methods in classifying imbalanced data sets on deep neural networks.
Wang, W, Yin, H, Sadiq, S, Chen, L, Xie, M & Zhou, X 1970, 'SPORE: A sequential personalized spatial item recommender system', 2016 IEEE 32nd International Conference on Data Engineering (ICDE), 2016 IEEE 32nd International Conference on Data Engineering (ICDE), IEEE, Helsinki, pp. 954-965.
View/Download from: Publisher's site
View description>>
© 2016 IEEE. With the rapid development of location-based social networks (LBSNs), spatial item recommendation has become an important way of helping users discover interesting locations to increase their engagement with location-based services. Although human movement exhibits sequential patterns in LBSNs, most current studies on spatial item recommendations do not consider the sequential influence of locations. Leveraging sequential patterns in spatial item recommendation is, however, very challenging, considering 1) users' check-in data in LBSNs has a low sampling rate in both space and time, which renders existing prediction techniques on GPS trajectories ineffective; 2) the prediction space is extremely large, with millions of distinct locations as the next prediction target, which impedes the application of classical Markov chain models; and 3) there is no existing framework that unifies users' personal interests and the sequential influence in a principled manner. In light of the above challenges, we propose a sequential personalized spatial item recommendation framework (SPORE) which introduces a novel latent variable topic-region to model and fuse sequential influence with personal interests in the latent and exponential space. The advantages of modeling the sequential effect at the topic-region level include a significantly reduced prediction space, an effective alleviation of data sparsity and a direct expression of the semantic meaning of users' spatial activities. Furthermore, we design an asymmetric Locality Sensitive Hashing (ALSH) technique to speed up the online top-k recommendation process by extending the traditional LSH. We evaluate the performance of SPORE on two real datasets and one large-scale synthetic dataset. The results demonstrate a significant improvement in SPORE's ability to recommend spatial items, in terms of both effectiveness and efficiency, compared with the state-of-the-art methods.
Wang, Z & Piccardi, M 1970, 'A pair hidden Markov support vector machine for alignment of human actions', 2016 IEEE International Conference on Multimedia and Expo (ICME), 2016 IEEE International Conference on Multimedia and Expo (ICME), IEEE, Seattle, WA, USA.
View/Download from: Publisher's site
View description>>
Alignment of human actions in videos is an important task for applications such as action comparison and classification. While well-established algorithms such as dynamic time warping are available for this task, they still heavily rely on basic linear cost models and heuristic parameter tuning. In this paper we propose a novel framework that combines the flexibility of the pair hidden Markov model (PHMM) with the effective parameter training of the structural support vector machine (SSVM). The framework extends the scoring function of SSVM to capture the similarity of two input sequences and introduces suitable feature and loss functions. The proposed approach is evaluated against state-of-the-art algorithms such as dynamic time warping (DTW) and canonical time warping (CTW) on pairs of human actions from the Weizmann and Olympic Sports datasets. The experimental results show that the proposed approach is capable of achieving an accuracy improvement of over 7 percentage points over the runner-up on both datasets.
WU, D, HUSSAIN, F, ZHANG, G, LU, JIE, UNWIN, J & RANCE, G 1970, 'A CLOUD-BASED COMPREHENSIVE HEALTH INFORMATION SYSTEM FRAMEWORK', Uncertainty Modelling in Knowledge Engineering and Decision Making, Conference on Uncertainty Modelling in Knowledge Engineering and Decision Making (FLINS 2016), WORLD SCIENTIFIC, pp. 612-617.
View/Download from: Publisher's site
View description>>
© 2016 by World Scientific Publishing Co. Pte. Ltd. Big data appearing in health domain bring great opportunities for the health information system development. To effectively utilize the big health data, three challenges: data heterogeneity, huge data volume and high velocity of data generation, and various kinds of user requirements, need to be dealt with. To solve the problem, this paper proposes a cloud-based comprehensive health information system framework, which uses cloud computing techniques to manage and process the big health data, and provides several data analysis and recommendation services to explore the data and extract values from them.
Wu, S, Jing, X-Y, Yue, D, Zhang, J, Yang, KJ & Yang, J 1970, 'Unsupervised visual domain adaptation via dictionary evolution', 2016 IEEE International Conference on Multimedia and Expo (ICME), 2016 IEEE International Conference on Multimedia and Expo (ICME), IEEE, Seattle, Washington, United States.
View/Download from: Publisher's site
View description>>
© 2016 IEEE. In real-word visual applications, distribution mismatch between samples from different domains may significantly degrade classification performance. To improve the generalization capability of classifier across domains, domain adaptation has attracted a lot of interest in computer vision. This work focuses on unsupervised domain adaptation which is still challenging because no labels are available in the target domain. Most of the attention has been dedicated to seeking domain-invariant feature by exploring the shared structure between domains, ignoring the valuable discriminative information contained in the labeled source data. In this paper, we propose a Dictionary Evolution (DE) approach to construct discriminative features robust to domain shift. Specifically, DE aims to adapt a discriminative dictionary learnt based on labeled source samples to unlabeled target samples through a gradual transition process. We show that the learnt dictionary is endowed with cross-domain data representation ability and powerful discriminant capability. Empirical results on real world data sets demonstrate the advantages of the proposed approach over competing methods.
Wu, W, Li, B, Chen, L & Zhang, C 1970, 'Canonical Consistent Weighted Sampling for Real-Value Weighted Min-Hash', 2016 IEEE 16th International Conference on Data Mining (ICDM), 2016 IEEE 16th International Conference on Data Mining (ICDM), IEEE, Barcelona, Spain, pp. 1287-1292.
View/Download from: Publisher's site
View description>>
© 2016 IEEE. Min-Hash, as a member of the Locality Sensitive Hashing (LSH) family for sketching sets, plays an important role in the big data era. It is widely used for efficiently estimating similarities of bag-of-words represented data and has been extended to dealing with multi-sets and real-value weighted sets. Improved ConsistentWeighted Sampling (ICWS) has been recognized as the state-of-The-Art for real-value weighted Min- Hash. However, the algorithmic implementation of ICWS is flawed because it violates the uniformity of the Min-Hash scheme. In this paper, we propose a Canonical Consistent Weighted Sampling (CCWS) algorithm, which not only retains the same theoretical complexity as ICWS but also strictly complies with the definition of Min-Hash. The experimental results demonstrate that the proposed CCWS algorithm runs faster than the state-of-The-Arts while achieving similar classification performance on a number of real-world text data sets.
Wu, W, Li, B, Chen, L & Zhang, C 1970, 'Cross-View Feature Hashing for Image Retrieval', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Pacific-Asia Conference on Knowledge Discovery and Data Mining, Springer International Publishing, Auckland, New Zealand, pp. 203-214.
View/Download from: Publisher's site
View description>>
© Springer International Publishing Switzerland 2016. Traditional cross-view information retrieval mainly rests on correlating two sets of features in different views. However, features in different views usually have different physical interpretations. It may be inappropriate to map multiple views of data onto a shared feature space and directly compare them. In this paper, we propose a simple yet effective Cross-View Feature Hashing (CVFH) algorithm via a “partition and match” approach. The feature space for each view is bi-partitioned multiple times using B hash functions and the resulting binary codes for all the views can thus be represented in a compatible B-bit Hamming space. To ensure that hashed feature space is effective for supporting generic machine learning and information retrieval functionalities, the hash functions are learned to satisfy two criteria: (1) the neighbors in the original feature spaces should be also close in the Hamming space; and (2) the binary codes for multiple views of the same sample should be similar in the shared Hamming space. We apply CVFH to cross view image retrieval. The experimental results show that CVFH can outperform the Canonical Component Analysis (CCA) based cross-view method.
Xie, J, Wang, M, Zhou, Y & Li, J 1970, 'Coordinating Discernibility and Independence Scores of Variables in a 2D Space for Efficient and Accurate Feature Selection', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Intelligent Computing Methodologies (ICIC), Springer International Publishing, Lanzhou, China, pp. 116-127.
View/Download from: Publisher's site
View description>>
© Springer International Publishing Switzerland 2016. Feature selection is to remove redundant and irrelevant features from original ones of exemplars, so that a sparse and representative feature subset can be detected for building a more efficient and accurate classifier. This paper presents a novel definition for the discernibility and independence scores of a feature, and then constructs a two dimensional (2D) space with the feature’s independence as y-axis and discernibility as x-axis to rank features’ importance. This new method is named FSDI (Feature Selection based on Discernibility and Independence of a feature). The discernibility score of a feature is to measure the distinguishability of the feature to detect instances from different classes. The independence score is to measure the redundancy of a feature. All features are plotted in the 2D space according to their discernibility and independence coordinates. The area of the rectangular corresponding to a feature’s discernibility and independence in the 2D space is used as a criterion to rank the importance of the features. Top-k features with much higher importance than the rest ones are selected to form the sparse and representative feature subset for building an efficient and accurate classifier. Experimental results on 5 classical gene expression datasets demonstrate that our proposed FSDI algorithm can select the gene subset efficiently and has the best performance in classification. Our method provides a good solution to the bottleneck issues related to the high time complexity of the existing gene subset selection algorithms.
Yang, D, Wu, Z, Wang, X, Cao, J & Xu, G 1970, 'Predicting Replacement of Smartphones with Mobile App Usage', Web Information Systems Engineering – WISE 2016 (LNCS), International Conference on Web Information Systems Engineering, Springer International Publishing, Shanghai, China, pp. 343-351.
View/Download from: Publisher's site
View description>>
© Springer International Publishing AG 2016.To identify right customers who intend to replace the smart phone can help to perform precision marketing and thus bring significant financial gains to cell phone retailers. In this paper,we provide a study of exploiting mobile app usage for predicting users who will change the phone in the future. We first analyze the characteristics of mobile log data and develop the temporal bag-of-apps model,which can transform the raw data to the app usage vectors. We then formularize the prediction problem,present the hazard based prediction model,and derive the inference procedure. Finally,we evaluate both data model and prediction model on real-world data. The experimental results show that the temporal usage data model can effectively capture the unique characteristics of mobile log data,and the hazard based prediction model is thus much more effective than traditional classification methods. Furthermore,the hazard model is explainable,that is,it can easily show how the replacement of smart phones relate to mobile app usage over time.
Yao, Y, Hua, X-S, Shen, F, Zhang, J & Tang, Z 1970, 'A Domain Robust Approach For Image Dataset Construction', Proceedings of the 24th ACM international conference on Multimedia, MM '16: ACM Multimedia Conference, ACM, Amsterdam, The Netherlands, pp. 212-216.
View/Download from: Publisher's site
View description>>
© 2016 ACM.There have been increasing research interests in automatically constructing image dataset by collecting images from the Internet. However, existing methods tend to have a weak domain adaptation ability, known as the \dataset bias problem". To address this issue, in this work, we propose a novel image dataset construction framework which can generalize well to unseen target domains. In specific, the given queries are first expanded by searching in the Google Books Ngrams Corpora (GBNC) to obtain a richer semantic description, from which the noisy query expansions are then filtered out. By treating each expansion as a \bag" and the retrieved images therein as \instances", we formulate image filtering as a multi-instance learning (MIL) problem with constrained positive bags. By this approach, images from different data distributions will be kept while with noisy images filtered out. Comprehensive experiments on two challenging tasks demonstrate the effectiveness of our proposed approach.
Yao, Y, Zhang, J, Hua, X-S, Shen, F & Tang, Z 1970, 'Extracting Visual Knowledge from the Internet: Making Sense of Image Data', MultiMedia Modeling (LNCS), International Conference on Multimedia Modeling, Springer International Publishing, Miami, USA, pp. 862-873.
View/Download from: Publisher's site
View description>>
© Springer International Publishing Switzerland 2016.Recent successes in visual recognition can be primarily attributed to feature representation, learning algorithms, and the everincreasing size of labeled training data. Extensive research has been devoted to the first two, but much less attention has been paid to the third. Due to the high cost of manual data labeling, the size of recent efforts such as ImageNet is still relatively small in respect to daily applications. In this work, we mainly focus on how to automatically generate identifying image data for a given visual concept on a vast scale. With the generated image data, we can train a robust recognition model for the given concept. We evaluate the proposed webly supervised approach on the benchmark Pascal VOC 2007 dataset and the results demonstrates the superiority of our method over many other state-ofthe- art methods in image data collection.
Yao, Y, Zhang, J, Shen, F, Hua, X, Xu, J & Tang, Z 1970, 'Automatic image dataset construction with multiple textual metadata', 2016 IEEE International Conference on Multimedia and Expo (ICME), 2016 IEEE International Conference on Multimedia and Expo (ICME), IEEE, Seattle, Washington, USA.
View/Download from: Publisher's site
View description>>
© 2016 IEEE. The goal of this work is to automatically collect a large number of highly relevant images from the Internet for given queries. A novel image dataset construction framework is proposed by employing multiple textual metadata. In specific, the given queries are first expanded by searching in the Google Books Ngrams Corpora to obtain a richer semantic description, from which the visually non-salient and less relevant expansions are then filtered. After retrieving images from the Internet with filtered expansions, we further filter noisy images by clustering and progressively Convolutional Neural Networks (CNN). To verify the effectiveness of our proposed method, we construct a dataset with 10 categories, which is not only much larger than but also have comparable cross-dataset generalization ability with manually labeled dataset STL-10 and CIFAR-10.
Ye, L, Cao, K, Guo, YJ, Huang, X, Beadle, P, Argha, A, Piccardi, M, Zhang, G & Su, SW 1970, 'Inertial Sensor based Post Fall Analysis for False Alarming Reduction', Telehealth and Assistive Technology / 847: Intelligent Systems and Robotics, Telehealth and Assistive Technology / 847: Intelligent Systems and Robotics, ACTAPRESS, Zurich, Switzerland, pp. 36-43.
View/Download from: Publisher's site
View description>>
One of the major public health problems among elderly people is falling injury. This study investigates fall detection and prevention by using inertial sensors for which the major existing challenging is how to significantly reduce false alarming in order to enhance the acceptance of elderly users during rehabilitation and daily exercises. Different from most existing approaches in the literature, the behavior after falling will be analyzed in details, which can not only greatly reduce false alarming, but also significantly improves the accuracy of the assessment of the severity of falling injuries.
Zeng, X, Lu, J, Kerre, EE, Martinez, L & Koehl, L 1970, 'Foreword', Uncertainty Modelling in Knowledge Engineering and Decision Making - Proceedings of the 12th International FLINS Conference, FLINS 2016, WORLD SCIENTIFIC PUBL CO PTE LTD, pp. v-vi.
View/Download from: Publisher's site
Zhang, D, Yin, J, Zhu, X & Zhang, C 1970, 'Homophily, Structure, and Content Augmented Network Representation Learning', 2016 IEEE 16th International Conference on Data Mining (ICDM), 2016 IEEE 16th International Conference on Data Mining (ICDM), IEEE, Barcelona, Spain, pp. 609-618.
View/Download from: Publisher's site
View description>>
© 2016 IEEE. Advances in social networking and communication technologies have witnessed an increasing number of applications where data is not only characterized by rich content information, but also connected with complex relationships representing social roles and dependencies between individuals. To enable knowledge discovery from such networked data, network representation learning (NRL) aims to learn vector representations for network nodes, such that off-The-shelf machine learning algorithms can be directly applied. To date, existing NRL methods either primarily focus on network structure or simply combine node content and topology for learning. We argue that in information networks, information is mainly originated from three sources: (1) homophily, (2) topology structure, and (3) node content. Homophily states social phenomenon where individuals sharing similar attributes (content) tend to be directly connected through local relational ties, while topology structure emphasizes more on global connections. To ensure effective network representation learning, we propose to augment three information sources into one learning objective function, so that the interplay roles between three parties are enforced by requiring the learned network representations (1) being consistent with node content and topology structure, and also (2) following the social homophily constraints in the learned space. Experiments on multi-class node classification demonstrate that the representations learned by the proposed method consistently outperform state-of-The-Art NRL methods, especially for very sparsely labeled networks.
Zhang, G & Piccardi, M 1970, 'Sequential Labeling with Structural SVM Under an Average Precision Loss', Structural, Syntactic, and Statistical Pattern Recognition, Joint IAPR International Workshop on Structural and Syntactic Pattern Recognition (SSPR) / International Workshop on Statistical Techniques in Pattern Recognition (SPR), Springer International Publishing, Mérida, Mexico, pp. 344-354.
View/Download from: Publisher's site
View description>>
© Springer International Publishing AG 2016.The average precision (AP) is an important and widelyadopted performance measure for information retrieval and classification systems. However, owing to its relatively complex formulation, very few approaches have been proposed to learn a classifier by maximising its average precision over a given training set. Moreover, most of the existing work is restricted to i.i.d. data and does not extend to sequential data. For this reason, we herewith propose a structural SVM learning algorithm for sequential labeling that maximises an average precision measure. A further contribution of this paper is an algorithm that computes the average precision of a sequential classifier at test time, making it possible to assess sequential labeling under this measure. Experimental results over challenging datasets which depict human actions in kitchen scenarios (i.e., TUM Kitchen and CMU Multimodal Activity) show that the proposed approach leads to an average precision improvement of up to 4.2 and 5.7% points against the runner-up, respectively.
Zhang, J, Zhang, J, Lu, J, Shen, C, Curr, K, Phua, R, Neville, R & Edmonds, E 1970, 'SLNSW-UTS: A Historical Image Dataset for Image Multi-Labeling and Retrieval', 2016 International Conference on Digital Image Computing: Techniques and Applications (DICTA), 2016 International Conference on Digital Image Computing: Techniques and Applications (DICTA), IEEE, Gold Coast, Australia, pp. 1-6.
View/Download from: Publisher's site
View description>>
© 2016 IEEE.This paper introduces a dataset of historical images created by the State Library of New South Wales and the University of Technology Sydney (UTS). The dataset has a total of 29713 images with 119 unique labels. Each image contains multiple labels. We use a CNN-based framework to explore the feasibility of our dataset in image multi-labeling and retrieval research, and extract semantic level image features for future research use. The experiment results illustrate that effective deep learning models can be trained on our dataset. We also introduce five applications that can be studied on our historical image dataset.
Zhang, Q, Wu, J, Yang, H, Lu, W, Long, G & Zhang, C 1970, 'Global and Local Influence-based Social Recommendation', Proceedings of the 25th ACM International on Conference on Information and Knowledge Management, CIKM'16: ACM Conference on Information and Knowledge Management, ACM, Indianapolis, USA, pp. 1917-1920.
View/Download from: Publisher's site
View description>>
© 2016 ACM. Social recommendation has been widely studied in recent years. Existing social recommendation models use various explicit pieces of social information as regularization terms, e.g., social links are considered as new constraints. However, social influence, an implicit source of information in social networks, is seldomly considered, even though it often drives recommendations in social networks. In this paper, we introduce a new global and local influence-based social recommendation model. Based on the observation that user purchase behaviour is influenced by both global influential nodes and the local influential nodes of the user, we formulate the global and local influence as an regularization terms, and incorporate them into a matrix factorization-based recommendation model. Experimental results on large data sets demonstrate the performance of the proposed method.
Zhang, Q, Wu, J, Zhang, P, Long, G, Tsang, IW & Zhang, C 1970, 'Inferring Latent Network from Cascade Data for Dynamic Social Recommendation', 2016 IEEE 16th International Conference on Data Mining (ICDM), 2016 IEEE 16th International Conference on Data Mining (ICDM), IEEE, Barcelona, Spain, pp. 669-678.
View/Download from: Publisher's site
View description>>
© 2016 IEEE. Social recommendation explores social information to improve the quality of a recommender system. It can be further divided into explicit and implicit social network recommendation. The former assumes the existence of explicit social connections between users in addition to the rating data. The latter one assumes the availability of only the ratings but not the social connections between users since the explicit social information data may not necessarily be available and usually are binary decision values (e.g., whether two people are friends), while the strength of their relationships is missing. Most of the works in this field use only rating data to infer the latent social networks. They ignore the dynamic nature of users that the preferences of users drift over time distinctly. To this end, we propose a new Implicit Dynamic Social Recommendation (IDSR) model, which infers latent social network from cascade data. It can sufficiently mine the information contained in time by mining the cascade data and identify the dynamic changes in the users in time by using the latest updated social network to make recommendations. Experiments and comparisons on three real-world datasets show that the proposed model outperforms the state-of-The-Art solutions in both explicit and implicit scenarios.
Zhang, Q, Zhang, Q, Long, G, Zhang, P & Zhang, C 1970, 'Exploring Heterogeneous Product Networks for Discovering Collective Marketing Hyping Behavior', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Pacific-Asia Conference on Knowledge Discovery and Data Mining, Springer International Publishing, Auckland, New Zealand, pp. 40-51.
View/Download from: Publisher's site
View description>>
© Springer International Publishing Switzerland 2016. Online spam comments often misguide users during online shopping. Existing online spam detection methods rely on semantic clues, behavioral footprints, and relational connections between users in review systems. Although these methods can successfully identify spam activities, evolving fraud strategies can successfully escape from the detection rules by purchasing positive comments from massive random users, i.e., user Cloud. In this paper, we study a new problem, Collective Marketing Hyping detection, for spam comments detection generated from the user Cloud. It is defined as detecting a group of marketing hyping products with untrustful marketing promotion behaviour. We propose a new learning model that uses heterogenous product networks extracted from product review systems. Our model aims to mining a group of hyping activities, which differs from existing models that only detect a single product with hyping activities. We show the existence of the Collective Marketing Hyping behavior in real-life networks. Experimental results demonstrate that the product information network can effectively detect fraud intentional product promotions.
Zhao, Y, Di, H, Zhang, J, Lu, Y & Lv, F 1970, 'Recognizing human actions from low-resolution videos by region-based mixture models', 2016 IEEE International Conference on Multimedia and Expo (ICME), 2016 IEEE International Conference on Multimedia and Expo (ICME), IEEE, Seattle, Washington, United States.
View/Download from: Publisher's site
View description>>
© 2016 IEEE. Recognizing human action from low-resolution (LR) videos is essential for many applications including large-scale video surveillance, sports video analysis and intelligent aerial vehicles. Currently, state-of-the-art performance in action recognition is achieved by the use of dense trajectories which are extracted by optical flow algorithms. However, the optical flow algorithms are far from perfect in LR videos. In addition, the spatial and temporal layout of features is a powerful cue for action discrimination. While, most existing methods encode the layout by previously segmenting body parts which is not feasible in LR videos. Addressing the problems, we adopt the Layered Elastic Motion Tracking (LEMT) method to extract a set of long-term motion trajectories and a long-term common shape from each video sequence, where the extracted trajectories are much denser than those of sparse interest points(SIPs); then we present a hybrid feature representation to integrate both of the shape and motion features; and finally we propose a Region-based Mixture Model (RMM) to be utilized for action classification. The RMM models the spatial layout of features without any needs of body parts segmentation. Experiments are conducted on two publicly available LR human action datasets. Among which, the UT-Tower dataset is very challenging because the average height of human figures is only about 20 pixels. The proposed approach attains near-perfect accuracy on both of the datasets.
Zheng, Y, Lan, C, Peng, H & Li, J 1970, 'Using constrained information entropy to detect rare adverse drug reactions from medical forums', 2016 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), 2016 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), IEEE, IEEE, pp. 2460-2463.
View/Download from: Publisher's site
View description>>
Adverse drug reactions (ADRs) detection is critical to avoid malpractices yet challenging due to its uncertainty in pre-marketing review and the underreporting in post-marketing surveillance. To conquer this predicament, social media based ADRs detection methods have been proposed recently. However, existing researches are mostly co-occurrence based methods and face several issues, in particularly, leaving out the rare ADRs and unable to distinguish irrelevant ADRs. In this work, we introduce a constrained information entropy (CIE) method to solve these problems. CIE first recognizes the drug-related adverse reactions using a predefined keyword dictionary and then captures high- and low-frequency (rare) ADRs by information entropy. Extensive experiments on medical forums dataset demonstrate that CIE outperforms the state-of-the-art co-occurrence based methods, especially in rare ADRs detection.
Zhou, T, Lu, Y, Di, H & Zhang, J 1970, 'Video object segmentation aggregation', 2016 IEEE International Conference on Multimedia and Expo (ICME), 2016 IEEE International Conference on Multimedia and Expo (ICME), IEEE, Seattle.
View/Download from: Publisher's site
View description>>
© 2016 IEEE. We present an approach for unsupervised object segmentation in unconstrained videos. Driven by the latest progress in this field, we argue that segmentation performance can be largely improved by aggregating the results generated by state-of-the-art algorithms. Initially, objects in individual frames are estimated through a per-frame aggregation procedure using majority voting. While this can predict relatively accurate object location, the initial estimation fails to cover the parts that are wrongly labeled by more than half of the algorithms. To address this, we build a holistic appearance model using non-local appearance cues by linear regression. Then, we integrate the appearance priors and spatio-temporal information into an energy minimization framework to refine the initial estimation. We evaluate our method on challenging benchmark videos and demonstrate that it outperforms state-of-the-art algorithms.
Zuo, Y, Wu, Q, An, P & Zhang, J 1970, 'Explicit measurement on depth-color inconsistency for depth completion', 2016 IEEE International Conference on Image Processing (ICIP), 2016 IEEE International Conference on Image Processing (ICIP), IEEE, Phoenix, AZ, USA, pp. 4037-4041.
View/Download from: Publisher's site
View description>>
© 2016 IEEE. Color-guided depth completion is to refine depth map through structure light sensing by filling missing depth structure and de-nosing. It is based on the assumption that depth discontinuity and color edge at the corresponding location are consistent. Among all proposed methods, MRF-based method including its variants is one of major approaches. However, the assumption above is not always true, which causes texture-copy and depth discontinuity blurring artifacts. The state-of-the-art solutions usually are to modify the weighting inside smoothness term of MRF model. Because there is no any method explicitly considering the inconsistency occurring between depth discontinuity and the corresponding color edge, they cannot adaptively control the effect of guidance from color image when completing depth map. In this paper, we propose quantitative measurement on such inconsistency and explicitly embed it into weighting value of smoothness term. The proposed method is evaluated on NYU Kinect datasets and demonstrates promising results.
Zuo, Y, Wu, Q, Zhang, J & An, P 1970, 'Explicit modeling on depth-color inconsistency for color-guided depth up-sampling', 2016 IEEE International Conference on Multimedia and Expo (ICME), 2016 IEEE International Conference on Multimedia and Expo (ICME), IEEE, USA.
View/Download from: Publisher's site
View description>>
© 2016 IEEE. Color-guided depth up-sampling is to enhance the resolution of depth map according to the assumption that the depth discontinuity and color image edge at the corresponding location are consistent. Through all methods reported, MRF including its variants is one of major approaches, which has dominated in this area for several years. However, the assumption above is not always true. Solution usually is to adjust the weighting inside smoothness term in MRF model. But there is no any method explicitly considering the inconsistency occurring between depth discontinuity and the corresponding color edge. In this paper, we propose quantitative measurement on such inconsistency and explicitly embed it into weighting value of smoothness term. Such solution has not been reported in the literature. The improved depth up-sampling based on the proposed method is evaluated on Middlebury datasets and ToFMark datasets and demonstrate promising results.
Zurita, G, Merigó, JM & Lobos-Ossandón, V 1970, 'A bibliometric analysis of journals in educational research', Lecture Notes in Engineering and Computer Science, pp. 403-408.
View description>>
The influence and impact of journals in the scientific community is a fundamental question for researchers worldwide because it measures the importance and quality of a publication. This study analyses all the journals that are currently ranked in any educational research category in Web of Science by using bibliometric indicators. The aim is to provide a general overview of their impact and influence between 1989 and 2013. The journals are divided in seven research categories that represent the whole field of educational research. The analysis also develops a general comparison between all the categories. The results show that many interdisciplinary journals obtain a broader impact than the core journals although these publications are also well positioned in the field.
Rawlings, J, Stoianoff, N, McCann, J, Catanaziti, T & Evers, M IP Australia 2016, 3.N. P. Stoianoff, E.Wright, J.McCann, J. Rawlings, T. Catanzariti and M. Evers, 2016, Submission on behalf of the Intellectual Property Program Educators and Researchers Faculty of Law, University of Technology Sydney to IP Australia regarding Exposure Draft regulations for the Trans-Tasman Patent Attorney Regime, June 2016., Sydney.
Rawlings, J, Stoianoff, N, McCann, J, Wilkinson, G, Wright, E & Cantanzariti, T The Productivity Commission 2016, 2.N.P. Stoianoff, J. McCann, J. Rawlings, G. Wilkinson, E. Wright, T. Catanzariti, 2016, Submission on behalf of the Intellectual Property Program Educators and Researchers, Faculty of Law UTS to The Productivity Commission regarding Intellectual Property Arrangements Draft Report April 2016, June 2016, 1-24., The Productivity Commission regarding Intellectual Property Arrangements Draft Report, pp. 1-24, Sydney.
Stoianoff, N, Wright, E, McCann, J, Rawlings, J, Catanzariti, T & Evers, M UTS 2016, Submission to IP Australia regarding the Exposure Draft of the Intellectual Property Legislation Amendment (Single Economic Market) Regulation 2016, Sydney.
Chalapathy, R, Borzeshi, EZ & Piccardi, M 2016, 'An Investigation of Recurrent Neural Architectures for Drug Name Recognition'.
Chalapathy, R, Borzeshi, EZ & Piccardi, M 2016, 'Bidirectional LSTM-CRF for Clinical Concept Extraction'.
Chalapathy, R, Borzeshi, EZ & Piccardi, M 2016, 'Bidirectional LSTM-CRF for Clinical Concept Extraction'.
Chang, X, Yang, Y, Long, G, Zhang, C & Hauptmann, AG 2016, 'Dynamic Concept Composition for Zero-Example Event Detection', arXiv.
Du, X, Yin, H, Chen, L, Wang, Y, Yang, Y & Zhou, X 2016, 'Personalized Video Recommendation Using Rich Contents from Videos'.
Han, B, Tsang, IW & Chen, L 2016, 'On the Convergence of A Family of Robust Losses for Stochastic Gradient Descent'.
Salvador, MM, Budka, M & Gabrys, B 2016, 'Automatic Composition and Optimization of Multicomponent Predictive Systems With an Extended Auto-WEKA'.
Stoianoff, N, Milne, J, Ashiabor, H, Kreiser, L & Butcher, B 2016, 'Market Instruments and the Protection of Natural Resources', Edward Elgar Publishing, pp. xii-xii.
View/Download from: Publisher's site
Zhang, W, Du, T & Wang, J 2016, 'Deep Learning over Multi-field Categorical Data: A Case Study on User Response Prediction'.
View description>>
Predicting user responses, such as click-through rate and conversion rate,
are critical in many web applications including web search, personalised
recommendation, and online advertising. Different from continuous raw features
that we usually found in the image and audio domains, the input features in web
space are always of multi-field and are mostly discrete and categorical while
their dependencies are little known. Major user response prediction models have
to either limit themselves to linear models or require manually building up
high-order combination features. The former loses the ability of exploring
feature interactions, while the latter results in a heavy computation in the
large feature space. To tackle the issue, we propose two novel models using
deep neural networks (DNNs) to automatically learn effective patterns from
categorical feature interactions and make predictions of users' ad clicks. To
get our DNNs efficiently work, we propose to leverage three feature
transformation methods, i.e., factorisation machines (FMs), restricted
Boltzmann machines (RBMs) and denoising auto-encoders (DAEs). This paper
presents the structure of our models and their efficient training algorithms.
The large-scale experiments with real-world data demonstrate that our methods
work better than major state-of-the-art models.