A class-dependent weighted dissimilarity measure for nearest neighbor classification problems

被引：52

作者：

Paredes, R ^{[1
]}

Vidal, E ^{[1
]}

机构：

[1] Univ Politecn Valencia, Inst Tecnol Informat, E-46071 Valencia, Spain

来源：

PATTERN RECOGNITION LETTERS | 2000年 / 21卷 / 12期

关键词：

nearest neighbour classification; weighted dissimilarity measures; iterative optimization; fractional programming;

D O I：

10.1016/S0167-8655(00)00064-7

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

A class-dependent weighted (CDW) dissimilarity measure in vector spaces is proposed to improve the performance of the nearest neighbor (NN) classifier. In order to optimize the required weights, an approach based on Fractional Programming is presented. Experiments with several standard benchmark data sets show the effectiveness of the proposed technique. (C) 2000 Published by Elsevier Science B.V.

引用

页码：1027 / 1036

页数：10

共 20 条

[11] SMALL SAMPLE-SIZE EFFECTS IN STATISTICAL PATTERN-RECOGNITION - RECOMMENDATIONS FOR PRACTITIONERS [J].

RAUDYS, SJ ;

JAIN, AK .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1991, 13 (03) :252-264

[12] THE OPTIMAL DISTANCE MEASURE FOR NEAREST NEIGHBOR CLASSIFICATION [J].

SHORT, RD ;

FUKUNAGA, K .

IEEE TRANSACTIONS ON INFORMATION THEORY, 1981, 27 (05) :622-627

[13]

SHORT RD, 1980, P 5 IEEE INT C PATT

[14] MODELING COGNITIVE-DEVELOPMENT ON BALANCE SCALE PHENOMENA [J].

SHULTZ, TR ;

MARESCHAL, D ;

SCHMIDT, WC .

MACHINE LEARNING, 1994, 16 (1-2) :57-86

[15]

SIGILLITO VG, 1989, J HOPKINS APL TECH D, V10, P262

[16]

SNIEDOVICH M, 1992, DYNAMIC PROGRAMMING

[17] GENERALIZATION OF K-NN RULE [J].

TOMEK, I .

IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS, 1976, 6 (02) :121-126

[18] GRADIENT DESCENT LEARNING OF NEAREST-NEIGHBOR CLASSIFIERS WITH OUTLIER REJECTION [J].

URAHAMA, K ;

FURUKAWA, Y .

PATTERN RECOGNITION, 1995, 28 (05) :761-768

[19] FAST COMPUTATION OF NORMALIZED EDIT DISTANCES [J].

VIDAL, E ;

MARZAL, A ;

AIBAR, P .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1995, 17 (09) :899-902

[20]

[No title captured]

← 1 2 →