Title :
A Method for Improving SVM Classifier by Excluding Redundant Information
Author :
Peng, Bing ; Zhou, Jianzhong ; Liu, Fang ; Fang, Rengcun
Abstract :
This paper proposes that support vectors include redundant information after analyzing Kernel´s geometrical structure and researching data dependant method for improving support vector machine (SVM). Redundant information confuses the law of a learning problem. Data dependant method on improving SVM is based on Riemannian geometry theory and could exclude redundant information. Reasoning and experiments show this method could effectively improve classification ability and classification speed of SVM.
Keywords :
geometry; pattern classification; support vector machines; Riemannian geometry theory; classification ability; data dependant method; redundant information; support vector machine; Data engineering; Educational institutions; Geometry; Hydroelectric power generation; Information analysis; Kernel; Machine learning; Risk management; Support vector machine classification; Support vector machines;
Conference_Titel :
Industrial Electronics and Applications, 2007. ICIEA 2007. 2nd IEEE Conference on
Conference_Location :
Harbin
Print_ISBN :
978-1-4244-0737-8
Electronic_ISBN :
978-1-4244-0737-8
DOI :
10.1109/ICIEA.2007.4318611
