Title :
Credit Risk Assessment with Least Squares Fuzzy Support Vector Machines
Author :
Yu, Lean ; Lai, Kin Keung ; Wang, Shouyang
Author_Institution :
Inst. of Syst. Sci., Chinese Acad. of Sci., Beijing
Abstract :
In this study, the authors discuss a least squares version of fuzzy support vector machine (FSVM) classifiers for designing a credit risk assessment system to discriminate good creditors from bad ones. Relative to the classical FSVM, the least squares FSVM (LS-FSVM) can transform a quadratic programming problem into a linear programming problem thus reducing the computational complexity. For illustration, a real-world credit dataset is used to test the effectiveness of the LS-FSVM
Keywords :
computational complexity; credit transactions; fuzzy reasoning; least squares approximations; linear programming; pattern classification; quadratic programming; risk management; support vector machines; FSVM; computational complexity; credit risk assessment; least squares fuzzy support vector machines; linear programming; quadratic programming; Artificial intelligence; Fuzzy systems; Least squares methods; Linear programming; Mathematics; Risk analysis; Risk management; Support vector machine classification; Support vector machines; Testing;
Conference_Titel :
Data Mining Workshops, 2006. ICDM Workshops 2006. Sixth IEEE International Conference on
Conference_Location :
Hong Kong
Print_ISBN :
0-7695-2702-7
DOI :
10.1109/ICDMW.2006.54
