Multi-Scale Adaptive Neighborhood Awareness Transformer For Graph Fraud Detection

2026-03-03

AI Summary

Graph neural networks for fraud detection are often hindered by the homophily assumption and insufficient global modeling capacity, limiting their ability to effectively distinguish between normal and fraudulent nodes. To address these challenges, this work proposes MANDATE (Multi-scale Neighborhood-aware Transformer), which enhances global context modeling through multi-scale positional encoding, introduces differentiated embedding strategies tailored to homophilic and heterophilic connections to mitigate distributional shift, and integrates multi-relational graph embeddings to reduce relational bias. Extensive experiments on three real-world fraud detection datasets demonstrate that MANDATE significantly outperforms existing methods, confirming its superior detection performance and generalization capability.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Social Networks and Social Media: Fairness and bias in social network and social media analysis
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Graph fraud detection (GFD) is crucial for identifying fraudulent behavior within graphs, benefiting various domains such as financial networks and social media. Existing methods based on graph neural networks (GNNs) have succeeded considerably due to their effective expressive capacity for graph-structured data. However, the inherent inductive bias of GNNs, including the homogeneity assumption and the limited global modeling ability, hinder the effectiveness of these models. To address these challenges, we propose Multi-scale Neighborhood Awareness Transformer (MANDATE), which alleviates the inherent inductive bias of GNNs. Specifically, we design a multi-scale positional encoding strategy to encode the positional information of various distances from the central node. By incorporating it with the self-attention mechanism, the global modeling ability can be enhanced significantly. Meanwhile, we design different embedding strategies for homophilic and heterophilic connections. This mitigates the homophily distribution differences between benign and fraudulent nodes. Moreover, an embedding fusion strategy is designed for multi-relation graphs, which alleviates the distribution bias caused by different relationships. Experiments on three fraud detection datasets demonstrate the superiority of MANDATE.

Problem

• Graph Fraud Detection
• Inductive Bias
• Homophily Assumption
• Global Modeling
• Graph Neural Networks

Innovation

• Multi-scale Positional Encoding
• Heterophilic-aware Embedding
• Graph Transformer
• Fraud Detection
• Embedding Fusion