Classification of Credit Applicants of Banking Systems Using Data Mining and Fuzzy Logic

This research study aims at using Data Mining and Fuzzy Logic approaches to classify the credit scoring of banking system applicants as to cover uncertainties and ambiguity connected with applicant classes and also variables that affect their behavior.The methodology, according to a standard Data Mining process, is to collect and refine the client data, then those variables which are in linguistic forms are converted to fuzzy variables under the supervision of banking experts and final data are modeled using Fuzzy Decision Tree, subsequently. The unfuzzy data are also modeled using the other algorithms.The results of the study suggest that as far as client distinction accuracy is concerned Fuzzy Decision Tree produces better results compared to Traditional Trees, Neural Networks, and statistical procedures such as Logistic Regression and Bayesian Network.However, it is not as accurate as Support Vector Machine and Genetic Tree. On the other hand, Fuzzy Decision Tree technique has gained better prediction than prediction performance of bank credit scoringexperts.