Previous [ 1] [ 2] [ 3] [ 4] [ 5] [ 6] [ 7] [ 8] [ 9] [ 10] [ 11] [ 12] [ 13] [ 14] [ 15] [ 16] [ 17] [ 18] [ 19] [ 20] [ 21] [ 22] [ 23] [ 24] [ 25]

@

Journal of Information Science and Engineering, Vol. 26 No. 6, pp. 2331-2339 (November 2010)

Support Vector Regression Based on Adjustable Entropy Function Approach*

QING WU+, SAN-YANG LIU AND LE-YOU ZHANG
+School of Automation
Xi'an Institute of Posts and Telecommunication
Xi'an, 710121 P.R. China
E-mail: xidianwq@yahoo.com.cn
Department of Mathematical Sciences
Xidian University
Xi'an, 710071 P.R. China

Support vector machine is an elegant tool for solving pattern recognition and regression problems. This paper presents a new smooth approach to solve support vector regression. Based on statistical learning theory and optimization theory, a smooth unconstrained optimization model for support vector regression is built with adjustable entropy technique. Newton descent method is used to solve the model. The proposed approach can overcome the numerical overflow in the traditional entropy function approaches. Primary numerical results illustrate that our proposed approach improves the regression performance and the learning efficiency.

Keywords: optimization, smooth technique, support vector regression, adjustable entropy function, Newton algorithm

Full Text () Retrieve PDF document (201011_25.pdf)

Received October 28, 2008; revised June 22 & September 21 & November 3, 2009; accepted December 21, 2009.
Communicated by Chin-Teng Lin.
* This paper was partially supported by the Nature Science Foundation of China (No.60674108 and 61075055), Natural Science Foundation of Shaanxi Province (No. 2010JQ8004) and Youth Foundation of Xian Institute of Posts and Telecommunications (No. 110-0402).