What do we Know About Fraud Detection in Peer-to-Peer Lending? A Systematic Literature Review

Abstract

This paper investigates fraud detection strategies in peer-to-peer (P2P) lending platforms, a key innovation in financial technology that faces significant fraud risks. Through a systematic literature review, we explore how P2P platforms can employ advanced analytics and machine learning models to detect and prevent fraud effectively. The study addresses two main questions: defining the types of fraud specific to P2P lending, such as identity theft and predatory lending, and examining the technologies for effectively detecting these fraudulent practices. Our findings emphasize the importance of real-time data analysis and the continuous updating of detection models to effectively identify and mitigate fraud risks. These strategies not only minimize financial losses but also enhance the security and trustworthiness of P2P lending platforms. The paper contributes to both practical applications and theoretical advancements in fraud detection, highlighting the need for robust frameworks to ensure a secure lending environment for all participants.

Keywords: Early Warning Systems, Customer Segmentation, Lending Settings, Unsupervised Learning, Systematic Literature Review