Застосування методів машинного навчання для кредитного скорінгу у випадку короткострокових позик

Abstract

The goal of the study is to determine a client's default risk level in the short term. This indicator, in turn, determines what kinds of loans the client can get. In this work, two machine learning methods, logistic regression and random forest, were used to process credit-based data and determine a client's default risk level in the short term. Logistic regression and random forest methods were chosen due to their effectiveness in handling large datasets and identifying relevant variables. The results obtained from the research can be used in banks and financial companies to improve the decision-making process and the efficiency of lending.