Title :
A new fuzzy support vector machines for class imbalance learning
Author :
Ma, Hongyan ; Wang, LiLing ; Shen, Bo
Author_Institution :
Ind. & Commercial Coll., Hebei Univ., Baoding, China
Abstract :
A bilateral-weighted fuzzy support vector machine (B-FSVM) proposed by Wang is to evaluate bank´s credit risk. However, it also suffers from the problem of class imbalance datasets in most cases. In this paper, we present a method to impve B-FSVM for class imbalance learning (called NFSVM-CIL) to handle the class imbalance problem in the presence of outliroers and noise. We evaluate and compare its performance with support vector machine, fuzzy support vector machine and FSVM-CIL.
Keywords :
banking; fuzzy set theory; learning (artificial intelligence); support vector machines; B-FSVM; banks credit risk evaluation; bilateral-weighted fuzzy support vector machine; imbalance learning class; Accuracy; Educational institutions; Learning systems; Machine learning; Noise; Support vector machines; Training; bilateral-weighted; class imbalance; fuzzy support vector machine; performance;
Conference_Titel :
Electrical and Control Engineering (ICECE), 2011 International Conference on
Conference_Location :
Yichang
Print_ISBN :
978-1-4244-8162-0
DOI :
10.1109/ICECENG.2011.6056838