Abstract

Evaluating in determining the eligibility of giving credit is very important. Errors in providing credit worthiness assessments can result a bad credit risk. The problem that often occurs is not the application of the system by financial parties but more on HR when making predictions about the determination of consumer credit worthiness. Research in the field of computers has been done to reduce credit risk resulting in losses to the company. In this research a comparison of Logistic Regression (LR), Naïve Bayes (NB) and Decision Tree (C4.5) algorithms is performed to predict the feasibility of granting credit. In order to produce a prediction of the feasibility of granting credit to new consumers, credit data used by the company is used. The data used in this study consists of 481 consumer records that have been classified as consumers with current credit and bad credit. After testing using the same dataset on the three algorithms by comparing the AUC and Confusion Matrix values, it was found that the appropriate algorithm to be applied to the credit worthiness dataset was Logistic Regression with an Area Under Curve (AUC) value of 0.972 and Accuracy or Confusion Matrix of 93.14%. As for the Decision Tree Algorithm (C4.5) from the test results, the AUC value is 0.926 and the Accuracy is 90.85% and the Algortima Naïve Bayes AUC value is 0.905 and the Accuracy is 82.75%.

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.