Enhancing Graph Neural Network-based Fraud Detectors against Camouflaged Fraudsters

Abstract

Graph Neural Networks (GNNs) have been widely applied to fraud detection problems in recent years, revealing the suspiciousness of nodes by aggregating their neighborhood information via different relations. However, few prior works have noticed the camouflage behavior of fraudsters, which could hamper the performance of GNN-based fraud detectors during the aggregation process. In this paper, we introduce two types of camouflages based on recent empirical studies, i.e., the feature camouflage and the relation camouflage. Existing GNNs have not addressed these two camouflages, which results in their poor performance in fraud detection problems. Alternatively, we propose a new model named CAmouflage-REsistant GNN (CARE-GNN), to enhance the GNN aggregation process with three unique modules against camouflages. Concretely, we first devise a label-aware similarity measure to find informative neighboring nodes. Then, we leverage reinforcement learning (RL) to find the optimal amounts of neighbors to be selected. Finally, the selected neighbors across different relations are aggregated together. Comprehensive experiments on two real-world fraud datasets demonstrate the effectiveness of the RL algorithm. The proposed CARE-GNN also outperforms state-of-the-art GNNs and GNN-based fraud detectors. We integrate all GNN-based fraud detectors as an opensource toolbox: https://github.com/safe-graph/DGFraud. The CARE-GNN code and datasets are available at https://github.com/YingtongDou/CARE-GNN.

arXiv commentAccepted by CIKM 2020

References in corpus (16)

• Semi-Supervised Classification with Graph Convolutional Networks
• Inductive Representation Learning on Large Graphs
• Accurate, Large Minibatch SGD: Training ImageNet in 1 Hour
• FastGCN: Fast Learning with Graph Convolutional Networks via Importance Sampling
• A Semi-supervised Graph Attentive Network for Financial Fraud Detection
• Alleviating the Inconsistency Problem of Applying Graph Neural Network to Fraud Detection
• Adversarial Attack and Defense on Graph Data: A Survey
• Spam Review Detection with Graph Convolutional Networks
• Learning Discrete Structures for Graph Neural Networks
• From Amateurs to Connoisseurs: Modeling the Evolution of User Expertise through Online Reviews
• Anti-Money Laundering in Bitcoin: Experimenting with Graph Convolutional Networks for Financial Forensics
• Adaptive Graph Convolutional Neural Networks
• Measuring and Improving the Use of Graph Information in Graph Neural Networks
• Deep Iterative and Adaptive Learning for Graph Neural Networks
• Secure Deep Graph Generation with Link Differential Privacy
• Friend or Faux: Graph-Based Early Detection of Fake Accounts on Social Networks

Cited by in corpus (53)

• A Comprehensive Survey on Graph Anomaly Detection with Deep Learning
• ECOD: Unsupervised Outlier Detection Using Empirical Cumulative Distribution Functions
• Federated Social Recommendation with Graph Neural Network
• ConsisRec: Enhancing GNN for Social Recommendation via Consistent Neighbor Aggregation
• Graph Neural Networks for Financial Fraud Detection: A Review
• EDITS: Modeling and Mitigating Data Bias for Graph Neural Networks
• Semi-supervised Credit Card Fraud Detection via Attribute-Driven Graph Representation
• Trustworthy Graph Neural Networks: Aspects, Methods and Trends
• GAD-NR: Graph Anomaly Detection via Neighborhood Reconstruction
• Alleviating Structural Distribution Shift in Graph Anomaly Detection
• RoSGAS: Adaptive Social Bot Detection with Reinforced Self-Supervised GNN Architecture Search
• Label Information Enhanced Fraud Detection against Low Homophily in Graphs
• ACE-HGNN: Adaptive Curvature Exploration Hyperbolic Graph Neural Network
• Graph Mining for Cybersecurity: A Survey
• New Benchmarks for Learning on Non-Homophilous Graphs
• Reinforced Neighborhood Selection Guided Multi-Relational Graph Neural Networks
• Action Sequence Augmentation for Early Graph-based Anomaly Detection
• Unsupervised Graph Outlier Detection: Problem Revisit, New Insight, and Superior Method
• Modeling User Behavior With Interaction Networks for Spam Detection
• Few-Shot Learning with Uncertainty-based Quadruplet Selection for Interference Classification in GNSS Data
• Class Label-aware Graph Anomaly Detection
• An Anomaly Event Detection Method Based on GNN Algorithm for Multi-data Sources
• Grad: Guided Relation Diffusion Generation for Graph Augmentation in Graph Fraud Detection
• Heterogeneous Similarity Graph Neural Network on Electronic Health Records
• Voucher Abuse Detection with Prompt-based Fine-tuning on Graph Neural Networks
• Few-shot Network Anomaly Detection via Cross-network Meta-learning
• 2SFGL: A Simple And Robust Protocol For Graph-Based Fraud Detection
• Label-Wise Graph Convolutional Network for Heterophilic Graphs
• Meta-GPS++: Enhancing Graph Meta-Learning with Contrastive Learning and Self-Training
• Basket Recommendation with Multi-Intent Translation Graph Neural Network
• Financial Crime & Fraud Detection Using Graph Computing: Application Considerations & Outlook
• UMGAD: Unsupervised Multiplex Graph Anomaly Detection
• EntropyStop: Unsupervised Deep Outlier Detection with Loss Entropy
• Dynamic Graph Collaborative Filtering
• Graph Computing for Financial Crime and Fraud Detection: Trends, Challenges and Outlook
• Relational Graph Neural Networks for Fraud Detection in a Super-App environment
• Multi-representations Space Separation based Graph-level Anomaly-aware Detection
• LPS-GNN : Deploying Graph Neural Networks on Graphs with 100-Billion Edges
• Polynomial Selection in Spectral Graph Neural Networks: An Error-Sum of Function Slices Approach
• Unlearning Inversion Attacks for Graph Neural Networks
• Learning Feature Encoder with Synthetic Anomalies for Weakly Supervised Graph Anomaly Detection
• Generative Dynamic Graph Representation Learning for Conspiracy Spoofing Detection
• Stable Prediction on Graphs with Agnostic Distribution Shift
• Relevance-Aware Anomalous Users Detection in Social Network via Graph Neural Network
• Random Similarity Isolation Forests
• Itsy Bitsy SpiderNet: Fully Connected Residual Network for Fraud Detection
• Partition-wise Graph Filtering: A Unified Perspective Through the Lens of Graph Coarsening
• Deep Fraud Detection on Non-attributed Graph
• Benchmarking Fraud Detectors on Private Graph Data
• Pre-training Recommender Systems via Reinforced Attentive Multi-relational Graph Neural Network
• EmoMed: An Emotionally-Aware Agent for Multimodal Medical Support with Real-Time Information Retrieval
• Temporal Graph Prototype-conditioned Conformal Prediction for Fraud Detection
• A Survey on Graph Neural Networks for Fraud Detection in Ride Hailing Platforms