A Multi-Perspective Fraud detection Method for Multi Participants E-Commerce Transactions

E-commerce platforms involve multiple participants including buyers, sellers, payment gateways, and logistics providers, each of which may become a potential target or source of fraudulent activities. Traditional fraud detection methods primarily focus on single-entity behavior, overlooking the collaborative and cross-participant dynamics that can lead to more complex fraud scenarios. To address this limitation, we propose a multi-perspective fraud detection method designed specifically for multi- participant e-commerce transactions. The approach integrates behavioral analytics, transaction pattern mining, and relationship modeling across participants to identify suspicious interactions in real time. By applying machine learning and graph-based anomaly detection techniques, the proposed model enhances fraud detection accuracy, minimizes false positives, and ensures greater trust in e-commerce ecosystems. Experimental evaluations on synthetic and real-world transaction datasets demonstrate the effectiveness of this method in detecting both individual and collusive fraud schemes, thereby contributing to the development of a secure and reliable e-commerce environment.