Some machine learning techniques for credit scoring

Abstract

Credit scoring is an important stage in the risk management process of financial institutions and banks. A good credit score will contribute to a better loan quality. Loan quality is the decisive aspect of the competition, existence and profitability of banks and financial institutions. Machine learning techniques have contributed to a significant improvement in the grading performance in credit scoring while also minimizing the risk of credit rating. The paper outlines some machine learning techniques for credit scoring that have been used by financial terms. The test results of machine learning techniques are obtained based on Kaggle datasets, the free datasets for data science research, which evaluate the classification efficiency of these machine learning techniques and suggest some recommendations. The deep learning algorithms give good results to study in the future.