Generative Graph Modelling for Anomaly Detection. This project aims to develop cutting-edge generative graph modelling techniques to address critical challenges in anomaly detection, including data im

Description

Generative Graph Modelling for Anomaly Detection. This project aims to develop cutting-edge generative graph modelling techniques to address critical challenges in anomaly detection, including data imbalance, lack of interpretability, and limited generalisability, for unforeseen anomalies. By leveraging diffusion dynamics to integrate local and global graph patterns, this project will discover hidden anomalies and provide human-understandable explanations. Anticipated outcomes include advanced generative graph algorithms, theories and models for next-generation anomaly detection. The research will benefit Australia’s economy and society, including advancements in fraud detection, smart devices, and urban transportation, while strengthening its global leadership in data science and AI.. Scheme: Discovery Projects. Field: 4605 - Data Management and Data Science. Lead: Prof Jia Wu