|
|
Please login using the form on menu list.
It is required to login for Full-Text PDF.
|
Distance between Two Classes: A Novel Kernel Class Separability Criterion
Jiancheng SUN
Chongxun ZHENG
Xiaohe LI
Publication
IEICE TRANSACTIONS on Information and Systems Vol.E92-D No.7 pp.1397-1400
Publication Date: 2009/07/01
Online ISSN: 1745-1361
Print ISSN: 0916-8532
Type of Manuscript: Special Section LETTER (Special Section on Large Scale Algorithms for Learning and Optimization)
Category:
Keyword:
kernel parameter,
data classification,
class separability,
support vector machine,
Full Text: PDF(131.9KB)
Summary:
With a Gaussian kernel function, we find that the distance between two classes (DBTC) can be used as a class separability criterion in feature space since the between-class separation and the within-class data distribution are taken into account impliedly. To test the validity of DBTC, we develop a method of tuning the kernel parameters in support vector machine (SVM) algorithm by maximizing the DBTC in feature space. Experimental results on the real-world data show that the proposed method consistently outperforms corresponding hyperparameters tuning methods.
|
|
