"THE MULTI-PERSPECTIVE FRAUD DETECTION METHOD FOR MULTI-PARTICIPANT FOR E-COMMERCE TRANSACTIONS"

E-commerce platforms have transformed global trade, but the increasing complexity of multi-participant transactions has led to rising concerns about fraudulent activities. Traditional fraud detection systems primarily focus on transactional features, often failing to capture the multi-faceted nature of fraudulent behavior in ecommerce ecosystems involving multiple stakeholders, including buyers, sellers, payment processors, and logistics providers. This study proposes a multi-perspective fraud detection framework that integrates behavioral analysis, transaction monitoring, and anomaly detection techniques. The system utilizes machine learning (ML) and deep learning models to analyze structured and unstructured data from different perspectives, including user behavior, transaction patterns, and external risk factors.