Machine learning in internet financial risk management: A systematic literature review
Xu Tian,ZongYi Tian,Saleh F. A. Khatib,Yan Wang
DOI: https://doi.org/10.1371/journal.pone.0300195
IF: 3.7
2024-04-16
PLoS ONE
Abstract:Internet finance has permeated into myriad households, bringing about lifestyle convenience alongside potential risks. Presently, internet finance enterprises are progressively adopting machine learning and other artificial intelligence methods for risk alertness. What is the current status of the application of various machine learning models and algorithms across different institutions? Is there an optimal machine learning algorithm suited for the majority of internet finance platforms and application scenarios? Scholars have embarked on a series of studies addressing these questions; however, the focus predominantly lies in comparing different algorithms within specific platforms and contexts, lacking a comprehensive discourse and summary on the utilization of machine learning in this domain. Thus, based on the data from Web of Science and Scopus databases, this paper conducts a systematic literature review on all aspects of machine learning in internet finance risk in recent years, based on publications trends, geographical distribution, literature focus, machine learning models and algorithms, and evaluations. The research reveals that machine learning, as a nascent technology, whether through basic algorithms or intricate algorithmic combinations, has made significant strides compared to traditional credit scoring methods in predicting accuracy, time efficiency, and robustness in internet finance risk management. Nonetheless, there exist noticeable disparities among different algorithms, and factors such as model structure, sample data, and parameter settings also influence prediction accuracy, although generally, updated algorithms tend to achieve higher accuracy. Consequently, there is no one-size-fits-all approach applicable to all platforms; each platform should enhance its machine learning models and algorithms based on its unique characteristics, data, and the development of AI technology, starting from key evaluation indicators to mitigate internet finance risks.
multidisciplinary sciences