Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
125 tokens/sec
GPT-4o
53 tokens/sec
Gemini 2.5 Pro Pro
42 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
47 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Revisiting Graph-Based Fraud Detection in Sight of Heterophily and Spectrum (2312.06441v3)

Published 11 Dec 2023 in cs.LG, cs.AI, and cs.SI

Abstract: Graph-based fraud detection (GFD) can be regarded as a challenging semi-supervised node binary classification task. In recent years, Graph Neural Networks (GNN) have been widely applied to GFD, characterizing the anomalous possibility of a node by aggregating neighbor information. However, fraud graphs are inherently heterophilic, thus most of GNNs perform poorly due to their assumption of homophily. In addition, due to the existence of heterophily and class imbalance problem, the existing models do not fully utilize the precious node label information. To address the above issues, this paper proposes a semi-supervised GNN-based fraud detector SEC-GFD. This detector includes a hybrid filtering module and a local environmental constraint module, the two modules are utilized to solve heterophily and label utilization problem respectively. The first module starts from the perspective of the spectral domain, and solves the heterophily problem to a certain extent. Specifically, it divides the spectrum into various mixed-frequency bands based on the correlation between spectrum energy distribution and heterophily. Then in order to make full use of the node label information, a local environmental constraint module is adaptively designed. The comprehensive experimental results on four real-world fraud detection datasets denote that SEC-GFD outperforms other competitive graph-based fraud detectors. We release our code at https://github.com/Sunxkissed/SEC-GFD.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (37)
  1. Graph based anomaly detection and description: a survey. Data mining and knowledge discovery, 29: 626–688.
  2. Beyond low-frequency information in graph convolutional networks. In AAAI, volume 35, 3950–3957.
  3. Measuring and relieving the over-smoothing problem for graph neural networks from the topological view. In AAAI, volume 34, 3438–3445.
  4. Adaptive universal generalized pagerank graph neural network. In ICLR.
  5. Efficient k-nearest neighbor graph construction for generic similarity measures. In WWW, 577–586.
  6. Enhancing graph neural network-based fraud detectors against camouflaged fraudsters. In CIKM, 315–324.
  7. Addressing heterophily in graph anomaly detection: A perspective of graph spectrum. In WWW, 1528–1538.
  8. Alleviating structural distribution shift in graph anomaly detection. In WSDM, 357–365.
  9. node2vec: Scalable feature learning for networks. In KDD, 855–864.
  10. Inductive representation learning on large graphs. In NeurIPS, 30.
  11. Revisiting graph neural networks: Graph filtering perspective. In ICPR, 8376–8383. IEEE.
  12. Graph-based fraud detection in the face of camouflage. TKDD, 11(4): 1–26.
  13. A GNN-based Few-shot learning model on the Credit Card Fraud detection. In DTPI, 320–323. IEEE.
  14. A Rayleigh quotient-based recursive total-least-squares online maximum capacity estimation for lithium-ion batteries. IEEE Transactions on Energy Conversion, 30(3): 842–851.
  15. Semi-supervised classification with graph convolutional networks. In ICLR.
  16. Pick and choose: a GNN-based imbalanced learning approach for fraud detection. In WWW, 3168–3177.
  17. Alleviating the inconsistency problem of applying graph neural network to fraud detection. In SIGIR, 1569–1572.
  18. Revisiting heterophily for graph neural networks. In NeurIPS, 35: 1362–1375.
  19. From amateurs to connoisseurs: modeling the evolution of user expertise through online reviews. In WWW, 897–908.
  20. Deepwalk: Online learning of social representations. In KDD, 701–710.
  21. Fraud detection: A systematic literature review of graph-based anomaly detection approaches. Decision Support Systems, 133: 113303.
  22. Explainable Graph-based Fraud Detection via Neural Meta-graph Search. In CIKM, 4414–4418.
  23. Collective opinion spam detection: Bridging review networks and metadata. In KDD, 985–994.
  24. H2-fdetector: A gnn-based fraud detector with homophilic and heterophilic connections. In WWW, 1486–1494.
  25. Rethinking graph neural networks for anomaly detection. In ICML, 21076–21089. PMLR.
  26. Multilayer perceptron (MLP). Geomatic approaches for modeling land change scenarios, 451–455.
  27. Graph attention networks. In ICLR.
  28. Deep structure learning for fraud detection. In ICDM, 567–576. IEEE.
  29. Label Information Enhanced Fraud Detection against Low Homophily in Graphs. In WWW, 406–416.
  30. Simplifying graph convolutional networks. In ICML, 6861–6871. PMLR.
  31. Semi-supervised Credit Card Fraud Detection via Attribute-Driven Graph Representation. In AAAI, volume 37, 14557–14565.
  32. How powerful are graph neural networks? In ICLR.
  33. Graph contrastive learning with augmentations. In NeurIPS, 33: 5812–5823.
  34. Fraudre: Fraud detection dual-resistant to graph inconsistency and imbalance. In ICDM, 867–876. IEEE.
  35. Temporal burstiness and collaborative camouflage aware fraud detection. Information Processing & Management, 60(2): 103170.
  36. An affect-rich neural conversational model with biased attention and weighted cross-entropy loss. In AAAI, volume 33, 7492–7500.
  37. Graph neural networks with heterophily. In AAAI, volume 35, 11168–11176.
Citations (10)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now