Journal
NEURAL COMPUTING & APPLICATIONS
Volume 22, Issue 7-8, Pages 1627-1635Publisher
SPRINGER LONDON LTD
DOI: 10.1007/s00521-012-0894-5
Keywords
Classification; Support vector machines; Twin support vector machines; Maximum margin; Unity norm hyperplanes
Categories
Funding
- National Natural Science Foundation of China [10971223, 10926198, 11071252]
- Zhejiang Provincial Natural Science Foundation of China [Y1100237]
Ask authors/readers for more resources
For classification problems, twin support vector machine (TWSVM) determines two nonparallel hyperplanes by solving two related SVM-type problems. TWSVM classifies binary patterns by the proximity of it to one of the two nonparallel hyperplanes. Thus, to calculate the distance of a pattern from the hyperplane, we need the unity norm of the normal vector of the hyperplane. But in the formulation of TWSVM, these equality constraints were not considered. In this paper, we consider unity norm constraints by using Euclidean norm and add a regularization term with the idea of maximizing some margin in TWSVM and propose a novel margin-based twin support vector machines with unity norm hyperplanes (UNH-MTSVM). We solved UNH-MTSVM by Newton's method, and the solution is updated by conjugate gradient method. The performance of both the linear and nonlinear UNH-MTSVM is verified experimentally on several bench mark and synthetic datasets. Experimental results show the effectiveness of our methods in both computation time and classification accuracy.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available