DocumentCode :
3635560
Title :
A 1-NN preclassifier for fuzzy k-NN rule
Author :
A. Jozwik;L. Chmieleski;W. Cudny;M. Sklodowski
Author_Institution :
Inst. of Biocybernetics & Biomed. Eng., Acad. of Sci., Warsaw, Poland
Volume :
4
fYear :
1996
Firstpage :
234
Abstract :
A two-stage classification scheme is presented. In the first stage it is decided which of the two rules, 1-NN or fuzzy k-NN, will be used in the second stage. The proposed approach leads to significant acceleration of the learning process as well as the classification phase.
Keywords :
"Neural networks","Error analysis","Cybernetics","Biomedical engineering","Image processing","Acceleration","Pattern recognition","Biomedical measurements","Probability","Costs"
Publisher :
ieee
Conference_Titel :
Pattern Recognition, 1996., Proceedings of the 13th International Conference on
ISSN :
1051-4651
Print_ISBN :
0-8186-7282-X
Type :
conf
DOI :
10.1109/ICPR.1996.547422
Filename :
547422
Link To Document :
بازگشت