The following information was submitted:
Transactions: WSEAS TRANSACTIONS ON MATHEMATICS
Transactions ID Number: 53-741
Full Name: Jing Yang
Position: Student
Age: ON
Sex: Female
Address: Liaocheng, Hunan Road No.1
Country: CHINA
Tel:
Tel prefix:
Fax:
E-mail address: yangjing860204@163.com
Other E-mails:
Title of the Paper: weighted generalized kernel discriminant analysis using fuzzy memberships
Authors as they appear in the Paper:
Email addresses of all the authors:
Number of paper pages: 12
Abstract: Linear discriminant analysis (LDA) is a classical approach for dimensionality reduction. However, LDA has limitations in that one of the scatter matrices is required to be nonsingular and the nonlinearly clustered structure is not easily captured. In order to overcome these problems, in this paper, we present several generalizations of kernel fuzzy discriminant analysis (KFDA) which include KFDA based on generalized singular value decomposition (KFDA/GSVD), pseudo-inverse KFDA (PIKFDA) and range space KFDA (RSKFDA). These KFDA-based algorithms adopts kernel methods to accommodate nonlinearly separable cases. In order to remedy the problem that KFDA-based algorithms fail to consider that different contribution of each pair of class to the discrimination, weighted schemes are incorporated into KFDA extensions in this paper and called them weighted generalized KFDA algorithms. Experiments on three real-world data sets are performed to test and evaluate the effectivene!
ss of the proposed algorithms and the effect of weights on classification accuracy. The results show that the effect of weighted schemes is very significantly.
Keywords: Kernel fuzzy discriminant analysis, Fuzzy membership, Undersampled problem, Weighting function, Classification accuracy
EXTENSION of the file: .pdf
Special (Invited) Session:
Organizer of the Session:
How Did you learn about congress:
IP ADDRESS: 113.125.238.155
