DocumentCode :
3216365
Title :
Kernel classification rules in the presence of missing values
Author :
Pawlak, Miroslaw ; Siedlecki, W.
Author_Institution :
Dept. of Electr. & Comput. Eng., Manitoba Univ., Winnipeg, Man., Canada
Volume :
i
fYear :
1990
fDate :
16-21 Jun 1990
Firstpage :
677
Abstract :
The nonparametric kernel classification rule derived from incomplete data is studied. Methods of designing kernel decision rules possessing optimal asymptotic properties are proposed. Consistency and rates of convergence are examined. It is argued that the replacement methods using the regression approach can lead to the inconsistency of resulting decision rules. On the other hand, a method employing the concept of predictive density yields asymptotically optimal classification rules
Keywords :
convergence; decision theory; nonparametric statistics; pattern recognition; consistency; kernel decision rules; nonparametric kernel classification rule; optimal asymptotic properties; predictive density; regression approach; Bismuth; Convergence; Design methodology; Equations; Kernel; Knowledge based systems; Linear regression; Process design; Regression analysis; Vectors;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Pattern Recognition, 1990. Proceedings., 10th International Conference on
Conference_Location :
Atlantic City, NJ
Print_ISBN :
0-8186-2062-5
Type :
conf
DOI :
10.1109/ICPR.1990.118190
Filename :
118190
Link To Document :
بازگشت