CREDIT SCORING BY ARTIFICIAL NEURAL NETWORKS BASED CROSS-ENTROPY AND FUZZY RELATIONS

Abstract

The credit scoring is one of the major activities in the banking sector. Because of growing market and increasing the loan applications, this field still continues its concern in terms of rating the applicants and assessing the credit amounts. To reduce the number of wrong decisions in the credit evaluation process, the decision makers focus on estimating more robust models. However, the traditional methods are criticized due to various pre-requisites and linear approximations in the high dimensional and excessive nonlinear cases. For this reason, artificial intelligence techniques are mostly preferred to handle the credit scoring problems accurately. This study presents an efficient procedure that is based on ANNs with cross-entropy and fuzzy relations in the context of the credit scoring. In the implementations, the proposed procedure is applied to a couple of benchmark credit scoring data sets and its performance is compared with traditional approaches.