Deep Learning Anti-Fraud Model for Internet Loan: Where We Are Going

Weiwei Fang,Xin Li,Ping Zhou,Jingwen Yan,Teng Zhou,Dazhi Jiang

doi:10.1109/access.2021.3051079

Weiwei Fang, Xin Li + Show 4 more

Open Access

https://doi.org/10.1109/access.2021.3051079

Copy DOI

Abstract

Recently, Internet finance is increasingly popular. However, bad debt has become a serious threat to Internet financial companies. The fraud detection models commonly used in conventional financial companies is logistic regression. Although it is interpretable, the accuracy of the logistic regression still remains to be improved. This paper takes a large public loan dataset, e.g. Lending club, for example, to explore the potential of applying deep neural network for fraud detection. We first fill the missing values by a random forest. Then, an XGBoost algorithm is employed to select the most discriminate features. After that, we propose to use a synthetic minority oversampling technique to deal with the sample imbalance. With the preprocessed data, we design a deep neural network for Internet loan fraud detection. Extensive experiments have been conducted to demonstrate the outperformance of the deep neural network compared with the commonly-used models. Such a simple yet effective model may brighten the application of deep learning in anti-fraud for Internet loans, which would benefit the financial engineers in small and medium Internet financial companies.

Highlights

Internet fraud methods are increasing dramatically in recent years, together with the rapid development of Internet financial models and the Internet business used to be handled by traditional financial institutions
PERFORMANCE EVALUATION AND DISCUSSIONS We evaluate the performance of the deep neural network by comparing with four commonly-used models, e.g. logistic regression (LS), support vector machine (SVM), decision tree (DT), and random forest (RF)
We introduce the main parameters of the model and optimizes to find the optimal parameter combination of the model

Summary

Introduction

Internet fraud methods are increasing dramatically in recent years, together with the rapid development of Internet financial models and the Internet business used to be handled by traditional financial institutions. In this regard, Internet lending companies face an unprecedented risk of online fraud. The rapid development of computer technology, the accumulating data, and the emerging data analysis techniques bring new opportunities to financial risk management and analysis on the big data in the financial industry. Researchers have developed various anti-fraud measures and fraud prevention systems over the years. Leonard [1] proposed a rule-based expert system for fraud detection.

Objectives

Methods

Findings

Conclusion