Weighted Parzen windows for pattern classification

被引:73
作者
Babich, GA [1 ]
Camps, OI [1 ]
机构
[1] PENN STATE UNIV, DEPT COMP SCI & ENGN, DEPT ELECT ENGN, UNIVERSITY PK, PA 16802 USA
关键词
nonparametric classifiers; Parzen-windows; kernel estimator; clustering; training samples; discriminant analysis; Bayes error; leave-one-out; holdout;
D O I
10.1109/34.494647
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
This correspondence introduces the weighted-Parzen-window classifier. The proposed technique uses a clustering procedure to find a set of reference Vectors and weights which are used to approximate the Parzen-window (kernel-estimator) classifier. The weighted-Parzen-window classifier requires less computation and storage than the full Parzen-window classifier. Experimental results showed that significant savings could be achieved with only minimal, if any, error rate degradation for synthetic and real data sets.
引用
收藏
页码:567 / 570
页数:4
相关论文
共 8 条
[1]  
ANDERBERG M, 1973, CLUSTER ANAL APPLICA
[2]  
[Anonymous], 1990, STAT PATTERN RECOGNI
[3]  
Duda R. O., 1973, PATTERN CLASSIFICATI, V3
[4]  
Fan J., 1994, Journal of computational and graphical statistics, V3, P35, DOI 10.2307/1390794
[5]   The use of multiple measurements in taxonomic problems [J].
Fisher, RA .
ANNALS OF EUGENICS, 1936, 7 :179-188
[6]  
Jain A. K., 1988, Pattern Recognition and Artificial Intelligence. Towards an Integration. Proceedings of an International Workshop, P211
[7]   ESTIMATION OF A PROBABILITY DENSITY-FUNCTION AND MODE [J].
PARZEN, E .
ANNALS OF MATHEMATICAL STATISTICS, 1962, 33 (03) :1065-&
[8]  
WEST M, 1993, J ROY STAT SOC B, V55, P409