PREDICTING CREDITWORTHINESS USING CREDIT SCORING USING MACHINE LEARNING METHODS

Karmazin Anton Richardovich

The study of creditworthiness using machine learning methods is not only of academic interest, but also has direct practical significance for financial institutions, insurance companies and other market participants. The use of advanced machine learning techniques provides an opportunity not only for more accurate forecasting, but also for deeper data analysis, which, in turn, helps identify early signs of financial risks. Current trends in scoring include the use of digital technologies, including artificial intelligence, machine learning, and automation of scoring processes. At the same time, an integrated approach combining subjective and statistical approaches to scoring, as well as taking into account specific factors of the factoring business, has become widespread. Thus, the use of machine learning methods in credit scoring can significantly improve the accuracy of predicting borrowers' creditworthiness. However, it is necessary to regularly update models and maintain a balance between the complexity of the model and its interpretability. Choosing the right algorithm and carefully preparing the data are key success factors in building an effective scoring system for predicting borrowers' creditworthiness.

Keywords:scoring; machine learning; lending; forecasting; financial organizations.