With the recent growth of the consumer credit market, the default index of financial institutions has also followed. How to effectively prevent and control the risk of credit default while expanding the loan business is an urgent issue for financial institutions to solve. Therefore, the establishment of a scientific and effective credit risk prediction model has important practical significance. This paper chooses to introduce and use four data mining models: Logistic Regression, Support Vector Machine, Random Forest and Light GBM to conduct empirical analysis on the historical loan data of a certain credit platform. The optimal credit default prediction model is selected based on various indicators, aiming to provide reliable theoretical and technical support for financial institutions.
Paper
Full text
Research on Credit Risk Forecast Model Based on Data Mining Technology
Semantic Scholar · Computer Science · 2022
Abstract
With the recent growth of the consumer credit market, the default index of financial institutions has also followed. How to effectively prevent and control the risk of credit default while expanding the loan business is an urgent issue for financial institutions to solve. Therefore, the establishment of a scientific and effective credit risk prediction model has important practical significance. This paper chooses to introduce and use four data mining models: Logistic Regression, Support Vector Machine, Random Forest and Light GBM to conduct empirical analysis on the historical loan data of a certain credit platform. The optimal credit default prediction model is selected based on various indicators, aiming to provide reliable theoretical and technical support for financial institutions.