CAMERA: Adapting to Semantic Camouflage in Unsupervised Text-Attributed Graph Fraud Detection

2026-05-19

AI Summary

This work addresses the challenge of fraudsters evading unsupervised textual attributed graph-based fraud detection through semantic disguise. To this end, we propose CAMERA, a Case-adaptive Multi-clue Expert framework that leverages a self-decoupled mixture-of-experts architecture to model diverse fraud cues. CAMERA adaptively integrates a node’s own features with its neighborhood information via a context-aware gating mechanism. Furthermore, by explicitly accounting for the rarity of fraudulent instances, we formulate an expert-level unsupervised one-class learning objective that relaxes conventional strong assumptions on structural and attribute distributions. Extensive experiments on four challenging datasets demonstrate that CAMERA significantly outperforms state-of-the-art methods in detecting fraud nodes concealed by semantic disguise.

Abstract

Text-attributed graph fraud detection (TAGFD) plays a critical role in preventing fraudulent activities on online social and e-commerce platforms. However, to evade detection, fraudsters continuously evolve their camouflaging strategies by deliberately mimicking textual responses of benign users, thereby concealing their malicious purposes. This phenomenon, referred to as semantic camouflage, fundamentally undermines commonly relied assumptions on how structural and attribute cues can be exploited to identify fraudsters, and makes it difficult to spot fraudsters with unsupervised TAGFD. To bridge the gaps, we propose a Case-Adaptive Multi-cue Expert fRAmework (CAMERA) for unsupervised TAGFD. CAMERA employs an ego-decoupled mixture-of-experts architecture, where each expert specializes in modeling a distinct type of fraud-indicative cue. A context-informed gating model is introduced to jointly consider the ego node representation and its local neighborhood context for adaptive integration of cues learned by different experts. Furthermore, CAMERA leverages the inherent rarity of fraudsters to support unsupervised one-class learning with expert-level objectives that encourage modeling dominant benign patterns, thereby enabling reliable unsupervised detection of camouflaged fraudsters. Experiments on 4 challenging datasets show that CAMERA consistently outperforms competitors, showing its effectiveness against semantically camouflaged fraudsters. Code available at https://github.com/CampanulaBells/CAMERA

Problem

• semantic camouflage
• unsupervised fraud detection
• text-attributed graph
• fraudster evasion
• graph-based anomaly detection

Innovation

• semantic camouflage
• unsupervised fraud detection
• mixture-of-experts
• text-attributed graph
• one-class learning