Unsupervised feature selection using a neuro-fuzzy approach

被引：73

作者：

Basak, J ^{[1
]}

De, RK ^{[1
]}

Pal, SK ^{[1
]}

机构：

[1] Indian Stat Inst, Machine Intelligence Unit, Calcutta 700035, W Bengal, India

来源：

PATTERN RECOGNITION LETTERS | 1998年 / 19卷 / 11期

关键词：

D O I：

10.1016/S0167-8655(98)00083-X

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

A neuro-fuzzy methodology is described which involves connectionist minimization of a fuzzy feature evaluation index with unsupervised training. The concept of a flexible membership function incorporating weighed distance is introduced in the evaluation index to make the modeling of clusters more appropriate. A set of optimal weighing coefficients in terms of networks parameters representing individual feature importance is obtained through connectionist minimization. Besides, the investigation includes the development of another algorithm for ranking of different feature subsets using the aforesaid fuzzy evaluation index without neural networks. Results demonstrating the effectiveness of the algorithms for various real life data are provided. (C) 1998 Published by Elsevier Science B.V. All rights reserved.

引用

页码：997 / 1006

页数：10

共 10 条

[1] PROTOTYPE CLASSIFICATION AND FEATURE SELECTION WITH FUZZY SETS
BEZDEK, JC
CASTELAZ, PF
[J]. IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS, 1977, 7 (02): : 87 - 92
[2] Feature analysis: Neural network and fuzzy set theoretic approaches
De, RK
Pal, NR
Pal, SK
[J]. PATTERN RECOGNITION, 1997, 30 (10) : 1579 - 1590
[3] The use of multiple measurements in taxonomic problems
Fisher, RA
[J]. ANNALS OF EUGENICS, 1936, 7 : 179 - 188
[4] HAYASHI Y, 1991, ADV NEURAL INFORMATI, P578
[5] FUZZY SET THEORETIC MEASURE FOR AUTOMATIC FEATURE EVALUATION
PAL, SK
CHAKRABORTY, B
[J]. IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS, 1986, 16 (05): : 754 - 760
[6] FUZZY SET THEORETIC MEASURES FOR AUTOMATIC FEATURE EVALUATION .2.
PAL, SK
[J]. INFORMATION SCIENCES, 1992, 64 (1-2) : 165 - 179
[7] Pal SK, 1986, Fuzzy mathematical approach to pattern recognition
[8] Automated feature selection with a distinction sensitive learning vector quantizer
Pregenzer, M
Pfurtscheller, G
Flotzinger, D
[J]. NEUROCOMPUTING, 1996, 11 (01) : 19 - 29
[9] BAYESIAN SELECTION OF IMPORTANT FEATURES FOR FEEDFORWARD NEURAL NETWORKS
PRIDDY, KL
ROGERS, SK
RUCK, DW
TARR, GL
KABRISKY, M
[J]. NEUROCOMPUTING, 1993, 5 (2-3) : 91 - 103
[10] Improved feature screening in feedforward neural networks
Steppe, JM
Bauer, KW
[J]. NEUROCOMPUTING, 1996, 13 (01) : 47 - 58

← 1 →