Researchers have introduced TSAI-MetaFraud, a new multimodal benchmark dataset designed to address the lack of comprehensive resources for detecting fraud and illicit behavior in virtual economies. Unlike existing datasets that analyze user behavior, authentication, or financial transactions in isolation, this resource integrates behavioral, transactional, and graph-structured information.
- The dataset incorporates realistic scenarios involving both human fraudsters and automated bots.
- It defines four specific benchmark tasks: transaction fraud detection, cross-modal node classification, temporal link prediction, and weakly supervised fraud detection.
- Baseline evaluations are provided using standard machine learning models and graph neural networks to facilitate reproducible research.
TSAI-MetaFraud aims to advance the development of trustworthy AI and multimodal learning techniques by providing a unified framework for analyzing relational structures within emerging metaverse ecosystems.