Evi-VN: Hard Region Guided Virtual Node Evidence Injection for GNN-Based Fraud Detection

Key points

• Online platforms contain growing numbers of bots, deceptive reviewers, and scam accounts that imitate legitimate users.1
• Across diverse GNNs, we observe overlapping errors on a shared hard region, suggesting the presence of latent fraud evidence that graph topologies and standard features fail to capture.1
• Fraud-specific GNNs can mitigate particular graph pathologies, yet they still make limited use of heterogeneous evidence such as structured records, text, images, and audio; uniform multimodal fusion may also disturb nodes already handled reliably by the graph.1
• To our knowledge, Evi-VN is the first graph fraud detection framework to use feature isolated evidence chains to correct hard regions shared across GNNs.1

Sources (1)

• [1]Evi-VN: Hard Region Guided Virtual Node Evidence Injection for GNN-Based Fraud Detection
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 8, 10:39 AM

  We propose Evi-VN to learn and correct these shared blind spots rather than build another fraud detector.

  Online platforms contain growing numbers of bots, deceptive reviewers, and scam accounts that imitate legitimate users.

Extractive summary: sentences quoted from the sources.