Transaction Fraud Detection via an Adaptive Graph Neural Network

TL;DR

This paper introduces ASA-GNN, an adaptive graph neural network that enhances transaction fraud detection by learning discriminative representations, filtering noisy data, and addressing fraudster camouflage, outperforming existing methods on real datasets.

Contribution

The paper proposes a novel ASA-GNN model with adaptive neighbor sampling and diversity metrics to improve fraud detection accuracy and robustness against fraudster camouflage.

Findings

ASA-GNN outperforms state-of-the-art methods on three real datasets.

The adaptive neighbor selection improves discriminative feature learning.

The diversity metric effectively mitigates over-smoothing and fraudster camouflage.

Abstract

Many machine learning methods have been proposed to achieve accurate transaction fraud detection, which is essential to the financial security of individuals and banks. However, most existing methods leverage original features only or require manual feature engineering. They lack the ability to learn discriminative representations from transaction data. Moreover, criminals often commit fraud by imitating cardholders' behaviors, which causes the poor performance of existing detection models. In this paper, we propose an Adaptive Sampling and Aggregation-based Graph Neural Network (ASA-GNN) that learns discriminative representations to improve the performance of transaction fraud detection. A neighbor sampling strategy is performed to filter noisy nodes and supplement information for fraudulent nodes. Specifically, we leverage cosine similarity and edge weights to adaptively select neighbors with similar behavior patterns for target nodes and then find multi-hop neighbors for fraudulent nodes. A neighbor diversity metric is designed by calculating the entropy among neighbors to tackle the camouflage issue of fraudsters and explicitly alleviate the over-smoothing phenomena. Extensive experiments on three real financial datasets demonstrate that the proposed method ASA-GNN outperforms state-of-the-art ones.