A Multiperspective Fraud Detection Method for Multi-Participant E-commerce Transactions

In the realm of e-commerce, where transactions involve multiple participants such as buyers, sellers, and intermediaries, the detection of fraudulent activities presents a significant challenge. To address this issue, our proposed method focuses on a Mult perspective approach aimed at enhancing fraud detection accuracy and efficiency. The first step involves the detection of user behaviors, wherein we leverage various techniques such as behavioral analysis and examination of transaction histories to gain insights into normal user behavior patterns. By understanding typical user interactions within the e-commerce ecosystem, we establish a baseline against which abnormal behaviors can be identified. Subsequently, we delve into the analysis of abnormalities for feature extraction. Utilizing sophisticated anomaly detection algorithms, we scrutinize transaction data to uncover irregular patterns indicative of potentially fraudulent activities. This process allows us to extract important features that serve as key indicators for fraud detection. Finally, we employ an ensemble classification model to implement our fraud detection mechanism, avoiding reliance on a specific algorithm. Instead, we leverage the strengths of ensemble algorithms, such as Random Forest, Gradient Boosting, or AdaBoost. By feeding the extracted features into the ensemble model, we train it to discern between legitimate and fraudulent behaviors in multiparticipant e-commerce transactions. Ensemble methods are particularly well-suited for this task due to their ability to handle high-dimensional data and capture complex decision boundaries through the combination of diverse base models.

Keywords: Multiparticipant E-commerce Transactions, Fraud Detection, User Behaviors, Abnormalities Analysis, Ensemble Classification Model, Random Forest, Gradient Boosting, AdaBoost