Pattern classification with genetic algorithms: Incorporation of chromosome differentiation

被引：10

作者：

Bandyopadhyay, S ^{[1
]}

Pal, SK ^{[1
]}

机构：

[1] INDIAN STAT INST,MACHINE INTELLIGENCE UNIT,CALCUTTA 700035,W BENGAL,INDIA

来源：

PATTERN RECOGNITION LETTERS | 1997年 / 18卷 / 02期

关键词：

nature's sexual differentiation; optimization; pattern recognition; restricted crossover; Schema theorem; speech recognition;

D O I：

10.1016/S0167-8655(97)00005-6

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

A new genetic search strategy involving chromosome differentiation into two classes and a restricted form of crossover operation is defined. Its application to multi-dimensional pattern recognition problems is studied. Superiority of the classifier is established for four sets of different artificial and real-life data. Schema analysis is provided. (C) 1997 Published by Elsevier Science B.V.

引用

页码：119 / 131

页数：13