Evaluation of Machine Learning Models for Credit Scoring

Authors

  • Ramesh Cheripelli
  • Kovvuri Ramya Sri

Abstract

One of the most important application of  machine learning is credit scoring.It is uber important for banks and financial institutions to develop credit card or loan services to compete with foreign capital and obtain profits, On the other side it is urgent to improve the ability to control credit risks. Customers are the valuable assets of any bank. The payment of timely bills is important for the running of banks. But if the customers do not pay on time, it may incur huge loss to any financial organization. In this paper we try to build several models which will predict the credit score of customers. Credit score is calculated on banking and finance datasets. To show the relation between attributes, the correlation matrix is generated. In the experimental part, the graphs are generated, which shows the contrast for better analysis. This paper predicts and proposes the factors or attributes which optimize the profits of any banking organization. The best model will be selected based on the accuracy, sensitivity and specificity values obtained.

Downloads

Published

2020-01-18

Issue

Section

Articles