Title :
Rank Space Diversity: A Diversity Measure of Base Kernel Matrices
Author :
Luo, Linkai ; Lin, Chengde ; Peng, Hong ; Zhou, Qifeng
Author_Institution :
Xiamen Univ., Xiamen
fDate :
May 30 2007-June 1 2007
Abstract :
This paper studies the diversity measure of base kernel matrices. First, rank space diversity is proposed as a diversity measure of base kernel matrices. Then, a rule for choosing base kernel matrices is deduced by this diversity measure. Last, our rule´s validation is claimed by some experiments on artificial data set and benchmark data set.
Keywords :
learning (artificial intelligence); matrix algebra; base kernel matrices; diversity measure; rank space diversity; Automatic control; Automation; Data mining; Extraterrestrial measurements; Kernel; Machine learning; Matrices; Support vector machines; Testing; Training data; diversity measure of base kernel matrices; learning kernel matrices; rank space diversity;
Conference_Titel :
Control and Automation, 2007. ICCA 2007. IEEE International Conference on
Conference_Location :
Guangzhou
Print_ISBN :
978-1-4244-0818-4
Electronic_ISBN :
978-1-4244-0818-4
DOI :
10.1109/ICCA.2007.4376629
