An Overview of SVM-Based Classification In Credit Scoring

Developing and expanding banking operations will work with an efficient system to improve the country s economy, and will bring financial institutions in a competitive environment. The existence of such a system that can help banks achieve their goals requires a variety of platforms. The purpose of credit scoring models is to predict the probability of non-repayment of credit by the customer or classification of credit applicants. The benefits of this approach can be to save time, save costs, eliminate personal judgment, and increase the accuracy of customer loan assessments. Different statistical methods have been used in the field of credit rating. Meanwhile, Support Vector Machine (SVM) has been one of the most used and popular methods. In this review, the general structure of a SVM-based credit scoring model is introduced and categorized. Then, the results of the research that have been studied using this method on the German and Australian dataset have been sorted and compared.

Keywords:

Credit Scoring , SVM , Classification , Feature selection