{"title":"Credit Risk Assessment in Commercial Banks Based on Fuzzy Support Vector Machines","authors":"Qifeng Zhou, Cheng-de Lin","doi":"10.1109/ICARCV.2006.345152","DOIUrl":null,"url":null,"abstract":"Credit risk assessment plays an important role in banks credit risk management. The objective of credit assessment is to decide credit ranks, which denote the capacity of enterprises to meet their financial commitments. Traditional \"one-versus-one\" approach has been commonly used in the multi-classification method based on support vector machine (SVM). Since SVM for pattern recognition is based on binary classification, there will be unclassifiable regions when extended to multi-classification problems. Focus on this problem, a new credit risk assessment model based on fuzzy SVM is introduced in this paper that can give a reasonable classification for unclassifiable examples. Experiment results show that the fuzzy SVM method provides a better performance in generalization ability and assessment accuracy than conventional one-versus-one multi-classification approach","PeriodicalId":415827,"journal":{"name":"2006 9th International Conference on Control, Automation, Robotics and Vision","volume":"1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2006-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2006 9th International Conference on Control, Automation, Robotics and Vision","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICARCV.2006.345152","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Credit risk assessment plays an important role in banks credit risk management. The objective of credit assessment is to decide credit ranks, which denote the capacity of enterprises to meet their financial commitments. Traditional "one-versus-one" approach has been commonly used in the multi-classification method based on support vector machine (SVM). Since SVM for pattern recognition is based on binary classification, there will be unclassifiable regions when extended to multi-classification problems. Focus on this problem, a new credit risk assessment model based on fuzzy SVM is introduced in this paper that can give a reasonable classification for unclassifiable examples. Experiment results show that the fuzzy SVM method provides a better performance in generalization ability and assessment accuracy than conventional one-versus-one multi-classification approach