Title :
A new credit scoring method based on improved fuzzy support vector machine
Author :
Tang, Bo ; Qiu, Saibing
Author_Institution :
Math. & Comput. Sci. Dept., Hunan City Univ., Yiyang, China
Abstract :
The techniques of credit scoring are the effective measure for credit risk management, and research on credit scoring in China is meaningful. This paper has put forward the new thinking of the model of setting up the risk and scoring with the fuzzy support vector machine algorithm. The empirical results show that the algorithm is very practical, and it has good prediction accuracy and anti-noise ability.
Keywords :
financial data processing; fuzzy set theory; risk management; support vector machines; SVM; anti-noise ability; credit risk management; credit scoring method; prediction accuracy; vector machine algorithm; Accuracy; Error analysis; Neural networks; Noise; Support vector machines; Training; credit scoring; fuzzy membership; fuzzy support vector machine;
Conference_Titel :
Computer Science and Automation Engineering (CSAE), 2012 IEEE International Conference on
Conference_Location :
Zhangjiajie
Print_ISBN :
978-1-4673-0088-9
DOI :
10.1109/CSAE.2012.6272911
