Prototype search for a nearest neighbor classifier by a genetic algorithm

Author

Kangas, Jari

Author_Institution

Nokia Res. Center, Tampere, Finland

fYear

1999

fDate

1999

Firstpage

117

Lastpage

121

Abstract

Deals with the problem of finding good prototypes for a condensed nearest-neighbor classifier for a character recognition system based on directional codes. A prototype search by a genetic algorithm which is capable of creating new prototypes is compared against a direct prototype selection algorithm. It is shown, in a leave-one-out experiment, that the prototypes found by the genetic algorithm give significantly smaller intra-class distances to class members and significantly larger normalized inter-class distances

Keywords

character recognition; genetic algorithms; pattern classification; search problems; character recognition system; condensed nearest-neighbor classifier; direct prototype selection algorithm; directional codes; genetic algorithm; intra-class distance; leave-one-out experiment; normalized inter-class distance; prototype search; Character recognition; Classification algorithms; Electronic mail; Genetic algorithms; Heuristic algorithms; Nearest neighbor searches; Optimization methods; Pattern recognition; Problem-solving; Prototypes;

fLanguage

English

Publisher

ieee

Conference_Titel

Computational Intelligence and Multimedia Applications, 1999. ICCIMA '99. Proceedings. Third International Conference on

Conference_Location

New Delhi

Print_ISBN

0-7695-0300-4

Type

conf

DOI

10.1109/ICCIMA.1999.798513

Filename

798513