Skip to main content

Robust Meta Learning for Risk-aware Recommender System

Project Member(s): Zhang, G., Zhang, Q., Xuan, J.

Funding or Partner Organisation: Australian Research Council (ARC Discovery Projects)
Australian Research Council (ARC Discovery Projects)

Start year: 2022

Summary: Recommender systems are the core of many online services but they are unbelievably vulnerable to potential risks like shilling attacks, privacy leaks, and unexpected changes. This project aims to develop new meta learning methods that make real-world recommender systems more robust to the risks and uncertain environments. The results would significantly improve robustness for personalised recommender systems under these risks, especially in e-Government and e-Learning applications, and bring economic and social benefit to all Australians. The anticipated outcomes will also advance machine learning knowledge with a new robust meta learning schema for general data analytics and recommender system with risk-aware recommendation methodologies.

FOR Codes: INFORMATION AND COMMUNICATION SERVICES, Application software packages, Expanding knowledge in the information and computing sciences, Machine learning, Artificial intelligence