An improved branch and bound algorithm for feature selection

被引：78

作者：

Chen, XW ^{[1
]}

机构：

[1] Calif State Univ Northridge, Dept Elect & Comp Engn, Northridge, CA 91330 USA

来源：

PATTERN RECOGNITION LETTERS | 2003年 / 24卷 / 12期

关键词：

branch and bound algorithm; solution tree; feature selection; classification;

D O I：

10.1016/S0167-8655(03)00020-5

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Feature selection plays an important role in pattern classification. In this paper, we present an improved branch and bound algorithm for optimal feature subset selection. This algorithm searches for an optimal solution in a large solution tree in an efficient manner by cutting unnecessary paths which are guaranteed not to contain the optimal solution. Our experimental results demonstrate the effectiveness of the new algorithm. (C) 2003 Elsevier Science B.V. All rights reserved.

引用

页码：1925 / 1933

页数：9

共 8 条

[1]

FREY PW, 1991, MACH LEARN, V6, P161, DOI 10.1023/A:1022606404104

[2]

Fukunaga K., 1992, INTRO STAT PATTERN R, V2nd

[3] Statistical pattern recognition: A review [J].

Jain, AK ;

Duin, RPW ;

Mao, JC .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2000, 22 (01) :4-37

[4]

NARENDRA P, 1977, IEEE T COMPUT, V26, P917, DOI 10.1109/TC.1977.1674939

[5] FLOATING SEARCH METHODS IN FEATURE-SELECTION [J].

PUDIL, P ;

NOVOVICOVA, J ;

KITTLER, J .

PATTERN RECOGNITION LETTERS, 1994, 15 (11) :1119-1125

[6]

SOMOL P, 2000, WORLD MULT SYST CYB, V7, P646

[7] MACHINE LEARNING TECHNIQUES TO DIAGNOSE BREAST-CANCER FROM IMAGE-PROCESSED NUCLEAR FEATURES OF FINE-NEEDLE ASPIRATES [J].

WOLBERG, WH ;

STREET, WN ;

MANGASARIAN, OL .

CANCER LETTERS, 1994, 77 (2-3) :163-171

[8] A MORE EFFICIENT BRANCH-AND-BOUND ALGORITHM FOR FEATURE-SELECTION [J].

YU, B ;

YUAN, BZ .

PATTERN RECOGNITION, 1993, 26 (06) :883-889

← 1 →