Domain Driven Classification of Customer Credit Data for Intelligent Credit Scoring using Fuzzy set and MC2

Credit scoring or credit risk assessment is an important research issue in the banking industry. The major challenge of credit scoring is to recruit the profitable customers by predicting the bankrupts. The credit scoring carried out by traditional data driven approaches resulted only in an imprecise solution. Also the domain-driven based multiple criteria and multiple constraint (MC2) level programming approach results only in a satisfying solution. In this paper, a fuzzy set based domain driven approach for classification of customer credit data has been provided. The multiple criteria and multiple constraint level programming are used for scoring the customers based on the classifier. The domain expertise knowledge is used for building the linear combinational sets of attributes for classification. This hybrid approach will identify the class of best, good, satisfactory, bad and worst customers. Experiments are based on publicly available datasets in the UCI Machine Learning Repository. KeywordsCredit scoring, classification, domain-driven approach, linear combination, fuzzy set, MC2.