SCFCRC: Simultaneously Counteract Feature Camouflage and Relation Camouflage for Fraud Detection

TL;DR

This paper introduces SCFCRC, a Transformer-based fraud detection method that simultaneously counters feature and relation camouflage strategies used by fraudsters, improving detection accuracy.

Contribution

The paper proposes a novel approach combining feature camouflage filtering and relation camouflage refinement using contrastive learning and Mixture-of-Experts networks.

Findings

Outperforms state-of-the-art baselines on benchmark datasets

Effectively mitigates feature and relation camouflage strategies

Enhances robustness with a new regularization method

Abstract

In fraud detection, fraudsters often interact with many benign users, camouflaging their features or relations to hide themselves. Most existing work concentrates solely on either feature camouflage or relation camouflage, or decoupling feature learning and relation learning to avoid the two camouflage from affecting each other. However, this inadvertently neglects the valuable information derived from features or relations, which could mutually enhance their adversarial camouflage strategies. In response to this gap, we propose SCFCRC, a Transformer-based fraud detector that Simultaneously Counteract Feature Camouflage and Relation Camouflage. SCFCRC consists of two components: Feature Camouflage Filter and Relation Camouflage Refiner. The feature camouflage filter utilizes pseudo labels generated through label propagation to train the filter and uses contrastive learning that combines instance-wise and prototype-wise to improve the quality of features. The relation camouflage refiner uses Mixture-of-Experts(MoE) network to disassemble the multi-relations graph into multiple substructures and divide and conquer them to mitigate the degradation of detection performance caused by relation camouflage. Furthermore, we introduce a regularization method for MoE to enhance the robustness of the model. Extensive experiments on two fraud detection benchmark datasets demonstrate that our method outperforms state-of-the-art baselines.