Truncated Affinity Maximization: One-class homophily modeling for graph anomaly detection

• Main Authors: QIAO, Hezhe, PANG, Guansong
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 2023
• Subjects: graph anomaly detection; local node affinity; Truncated Affinity Maximization; Artificial Intelligence and Robotics
• Online Access: https://ink.library.smu.edu.sg/sis_research/8455; https://ink.library.smu.edu.sg/context/sis_research/article/9458/viewcontent/9585_truncated_affinity_maximizatio.pdf
• Institution: Singapore Management University

We reveal a one-class homophily phenomenon, which is one prevalent property we find empirically in real-world graph anomaly detection (GAD) datasets, i.e., normal nodes tend to have strong connection/affinity with each other, while the homophily in abnormal nodes is significantly weaker than normal...