Machine Learning for Accurate Air Pollution Forecast

Project Member(s): Ha, Q.

Funding or Partner Organisation: NSW Department of Planning, Industry and Environment
NSW Department of Planning, Industry and Environment

Start year: 2021

Summary: Upon recent results from application of LSTM on the data sets provided by DPIE, improvements could be reached by using the Bayesian neural networks (BNN) owing to the capability of inferring the distribution of model’s parameters (weights, bias, etc.) based on the new data given to the model. Here, this parameter distribution marginalisation contributes to the prediction making. So, a well-developed algorithm integrating long short-term memory with Bayesian neural network (LSTM-BNN) could allow for a comparative mix of different models, which is promising for robust air quality monitoring. Given the ultimate goal of improving forecast accuracy, this project is aimed at exploring, in terms of methodology and development, the best fit model for this purpose

FOR Codes: Environment, Automation engineering, ENVIRONMENTAL POLICY, CLIMATE CHANGE AND NATURAL HAZARDS, ENVIRONMENTAL MANAGEMENT

Acknowledgement of Country

UTS acknowledges the Gadigal people of the Eora Nation, the Boorooberongal people of the Dharug Nation, the Bidiagal people and the Gamaygal people, upon whose ancestral lands our university stands. We would also like to pay respect to the Elders both past and present, acknowledging them as the traditional custodians of knowledge for these lands.