Paper: arXiv 2502.19305

Abstract

Corporate fraud detection aims to automatically recognize companies that conduct wrongful activities such as fraudulent financial statements or illegal insider trading. Previous learning-based methods fail to effectively integrate rich interactions in the company network. To close this gap, we collect 18-year financial records in China to form three graph datasets with fraud labels. We analyze the characteristics of the financial graphs, highlighting two pronounced issues: (1) information overload: the dominance of (noisy) non-company nodes over company nodes hinders the message-passing process in Graph Convolution Networks (GCN); and (2) hidden fraud: there exists a large percentage of possible undetected violations in the collected data. The hidden fraud problem will introduce noisy labels in the training dataset and compromise fraud detection results. To handle such challenges, we propose a novel graph-based method, namely, Knowledge-enhanced GCN with Robust Two-stage Learning (${"\rm KeGCN"}{“R”}$), which leverages Knowledge Graph Embeddings to mitigate the information overload and effectively learns rich representations. The proposed model adopts a two-stage learning method to enhance robustness against hidden frauds. Extensive experimental results not only confirm the importance of interactions but also show the superiority of ${"\rm KeGCN"}{“R”}$ over a number of strong baselines in terms of fraud detection effectiveness and robustness.

Complexity vs Empirical Score

  • Math Complexity: 6.5/10
  • Empirical Rigor: 8.0/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper introduces advanced graph learning techniques (KeGCNR) with mathematical formulations for knowledge graph embeddings and robust two-stage learning, while providing extensive empirical validation on large, real-world financial datasets with public code and detailed experimental results.

Research Flowchart

  flowchart TD
  A["Research Goal: Integrate Rich Interactions<br>for Corporate Fraud Detection"] --> B["Key Inputs: 18-year Financial Graph Datasets<br>(3 Graphs with Fraud Labels)"]
  B --> C["Challenge Analysis: Financial Graph Characteristics"]
  C --> D["Challenge 1: Information Overload<br>(Non-company nodes dominate)"]
  C --> E["Challenge 2: Hidden Fraud<br>(Noisy training labels)"]
  D --> F["Method: KeGCN_R Model<br>Knowledge Graph Embedding + Robust Two-stage Learning"]
  E --> F
  F --> G["Computational Process:<br>1. Knowledge-enhanced GCN<br>2. Stage 1: Representation Learning<br>3. Stage 2: Robust Loss Optimization"]
  G --> H["Key Findings & Outcomes"]
  H --> I["1. Interactions are crucial for fraud detection"]
  H --> J["2. KeGCN_R outperforms strong baselines<br>(Higher Effectiveness & Robustness)"]

Cited References (90)

  1. Heterogeneous graph transformer (2020)
  2. Multiplex heterogeneous graph convolutional network (2022)
  3. Financial fraud detection using the related-party transaction knowledge graph (2022)
  4. Higher-order attribute-enhancing heterogeneous graph neural networks (2021)
  5. Graph Transformer Networks: Learning meta-path graphs to improve GNNs (2022)
  6. Graph transformer networks (2019)
  7. Application of machine learning methods to risk assessment of financial statement fraud: evidence from China (2014)
  8. Mining corporate annual reports for intelligent detection of financial statement fraud–a comparative study of machine learning methods (2017)
and 82 more
  1. Financial fraud detection model: based on random forest (2015)
  2. Deep learning for detecting financial statement fraud (2020)
  3. Intelligent financial fraud detection: a comprehensive review (2016)
  4. Detecting financial misstatements with fraud intention using multi-class cost-sensitive learning (2016)
  5. Detecting fraudulent financial reporting through financial statement analysis (2014)
  6. Detecting the financial statement fraud: the analysis of the differences between data mining techniques and experts’ judgments (2015)
  7. The detection of earnings manipulation (1999)
  8. Predicting material accounting misstatements (2011)
  9. Board composition and corporate fraud (2004)
  10. CEO connectedness and corporate fraud (2015)
  11. The benefits and costs of group affiliation: evidence from east Asia (2006)
  12. Propping through related party transactions (2010)
  13. Knowledge graph embedding for link prediction: a comparative analysis (2021)
  14. Knowledge graph embedding: a survey of approaches and applications (2017)
  15. Translating embeddings for modeling multi-relational data (2013)
  16. Complex embeddings for simple link prediction (2016)
  17. Layer normalization (2016) — link
  18. A three-way model for collective learning on multi-relational data (2011)
  19. Knowledge graph convolutional networks for recommender systems (2019)
  20. Learning entity and relation embeddings for knowledge graph completion (2015)
  21. Knowledge graph embedding by translating on hyperplanes (2014)
  22. Convolutional 2d knowledge graph embeddings (2018)
  23. Holographic embeddings of knowledge graphs (2016)
  24. DGL-KE: training knowledge graph embeddings at scale (2020)
  25. Kgat: knowledge graph attention network for recommendation (2019)
  26. Geometric deep learning: going beyond euclidean data (2017)
  27. Deeper insights into graph convolutional networks for semi-supervised learning (2018)
  28. A comprehensive survey on graph neural networks (2020)
  29. Semi-supervised classification with graph convolutional networks (2016) — link
  30. Graph Attention Networks (2018)
  31. Convolutional neural networks on graphs with fast localized spectral filtering (2016)
  32. Live-Streaming Fraud Detection: A Heterogeneous Graph Neural Network Approach (2021)
  33. Intention-aware heterogeneous graph attention networks for fraud transactions detection (2021)
  34. Improving graph-based label propagation algorithm with group partition for fraud detection (2020)
  35. ANOMALOUS: a joint modeling approach for anomaly detection on attributed networks. (2018)
  36. Towards consumer loan fraud detection: graph neural networks with role-constrained conditional random field (2021)
  37. Cash-out user detection based on attributed heterogeneous information network with a hierarchical attention mechanism (2019)
  38. Heterogeneous mini-graph neural network and its application to fraud invitation detection (2020)
  39. Heterogeneous graph attention network (2019)
  40. Heterogeneous graph neural network (2019)
  41. A survey on heterogeneous graph embedding: methods, techniques, applications and sources (2022)
  42. Magnn: Metapath aggregated graph neural network for heterogeneous graph embedding (2020)
  43. Pathsim: meta path-based top-k similarity search in heterogeneous information networks (2011)
  44. Knowledge embedding based graph convolutional network (2021)
  45. End-to-end structure-aware convolutional networks for knowledge base completion (2019)
  46. Modeling relational data with graph convolutional networks (2018)
  47. Xgboost: a scalable tree boosting system (2016)
  48. Random forests (2001)
  49. Least squares support vector machine classifiers (1999)
  50. Embedding entities and relations for learning and inference in knowledge bases (2015)
  51. RotatE: knowledge graph embedding by relational rotation in complex space (2018)
  52. Dimensionality reduction by learning an invariant mapping (2006)
  53. The case analysis of the scandal of Enron (2010)
  54. A second-order approach to learning with instance-dependent label noise (2021)
  55. Learning with instance-dependent label noise: a sample sieve approach (2020)
  56. Estimating instance-dependent Bayes-label transition matrix using a deep neural network (2022)
  57. Making deep neural networks robust to label noise: A loss correction approach (2017)
  58. A Model-Agnostic approach for learning with noisy labels of arbitrary distributions (2022)
  59. Towards deeper graph neural networks (2020)
  60. Robust training of graph neural networks via noise governance (2023)
  61. Nrgnn: Learning a label noise resistant graph neural network on sparsely and noisily labeled graphs. (2021)
  62. Inductive representation learning on large graphs (2017)
  63. Fraud detection under multi-sourced extremely noisy annotations (2021)
  64. On the minimal supervision for training any binary classifier from only unlabeled data (2018)
  65. Learning with bounded instance and label-dependent label noise (2020)
  66. Searching to exploit memorization effect in learning with noisy labels (2020)
  67. Detecting Cash-out users via dense subgraphs (2022)
  68. Company-as-Tribe: company financial risk assessment on Tribe-Style graph with hierarchical graph neural networks (2022)
  69. Information transmission between the NASDAQ and Asian second board markets (2004)
  70. A comparative study of the effects of different factors on firm technological innovation performance in different high-tech industries (2019)
  71. Learning from massive noisy labeled data for image classification (2015)
  72. Learning with noisy labels (2013)
  73. Noise tolerance under risk minimization (2013)
  74. Classification with noisy labels by importance reweighting (2015)
  75. Dual t: Reducing estimation error for transition matrix in label-noise learning (2020)
  76. User behavior pre-training for online fraud detection (2022)
  77. Part-dependent label noise: Towards instance-dependent label noise (2020)
  78. Deep learning for credit card fraud detection: A review of algorithms, challenges, and solutions (2024)
  79. Graph contrastive multi-view learning: A pre-training framework for graph classification (2024)
  80. Financial fraud detection using graph neural networks: A systematic review (2024)
  81. Graph Relearn Network: Reducing performance variance and improving prediction accuracy of graph neural networks (2024)
  82. Cost-Sensitive GNN-Based Imbalanced learning for mobile social network fraud detection (2023)

0 of 90 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2518 linked papers.