Comparison of feature selection algorithms in the context of rough classifiers

S. K. Choubey, Jitender S Deogun, V. V. Raghavan, H. Sever

Research output: Contribution to conferencePaper

29 Citations (Scopus)

Abstract

In this paper, we study the feature selection problem and develop and analyze four algorithms for feature selection in the context of rough set methodology. The initial state and the feasibility criterion of all these algorithms are the same, that is, they start from a given feature set and progressively remove features, while controlling the amount of degradation in classification quality, but differ in the heuristic used for pruning the search space of features. Our experimental results confirm the analytical results on the complexity of algorithms as well as on controlled degradation of upper classification. The algorithms presented can be used with any methods of deriving a classifier where the quality of classification is monotonically decreasing function while feature set is reduced, though we have adopted the upper classifier in our study. The upper classifier has some important features that makes it suitable for database mining applications. In particular, we have shown that the upper classifier can be summarized at a desired level of abstraction by using extended decision tables. We also point out that an inconsistent decision algorithm can be interpreted as if it were a consistent decision algorithm.

Original languageEnglish (US)
Pages1122-1128
Number of pages7
StatePublished - Dec 1 1996
EventProceedings of the 1996 5th IEEE International Conference on Fuzzy Systems. Part 3 (of 3) - New Orleans, LA, USA
Duration: Sep 8 1996Sep 11 1996

Other

OtherProceedings of the 1996 5th IEEE International Conference on Fuzzy Systems. Part 3 (of 3)
CityNew Orleans, LA, USA
Period9/8/969/11/96

Fingerprint

Feature extraction
Classifiers
Decision tables
Degradation

ASJC Scopus subject areas

  • Chemical Health and Safety
  • Software
  • Safety, Risk, Reliability and Quality

Cite this

Choubey, S. K., Deogun, J. S., Raghavan, V. V., & Sever, H. (1996). Comparison of feature selection algorithms in the context of rough classifiers. 1122-1128. Paper presented at Proceedings of the 1996 5th IEEE International Conference on Fuzzy Systems. Part 3 (of 3), New Orleans, LA, USA, .

Comparison of feature selection algorithms in the context of rough classifiers. / Choubey, S. K.; Deogun, Jitender S; Raghavan, V. V.; Sever, H.

1996. 1122-1128 Paper presented at Proceedings of the 1996 5th IEEE International Conference on Fuzzy Systems. Part 3 (of 3), New Orleans, LA, USA, .

Research output: Contribution to conferencePaper

Choubey, SK, Deogun, JS, Raghavan, VV & Sever, H 1996, 'Comparison of feature selection algorithms in the context of rough classifiers' Paper presented at Proceedings of the 1996 5th IEEE International Conference on Fuzzy Systems. Part 3 (of 3), New Orleans, LA, USA, 9/8/96 - 9/11/96, pp. 1122-1128.
Choubey SK, Deogun JS, Raghavan VV, Sever H. Comparison of feature selection algorithms in the context of rough classifiers. 1996. Paper presented at Proceedings of the 1996 5th IEEE International Conference on Fuzzy Systems. Part 3 (of 3), New Orleans, LA, USA, .
Choubey, S. K. ; Deogun, Jitender S ; Raghavan, V. V. ; Sever, H. / Comparison of feature selection algorithms in the context of rough classifiers. Paper presented at Proceedings of the 1996 5th IEEE International Conference on Fuzzy Systems. Part 3 (of 3), New Orleans, LA, USA, .7 p.
@conference{cc381738b50d43dda33c84dbde51f57a,
title = "Comparison of feature selection algorithms in the context of rough classifiers",
abstract = "In this paper, we study the feature selection problem and develop and analyze four algorithms for feature selection in the context of rough set methodology. The initial state and the feasibility criterion of all these algorithms are the same, that is, they start from a given feature set and progressively remove features, while controlling the amount of degradation in classification quality, but differ in the heuristic used for pruning the search space of features. Our experimental results confirm the analytical results on the complexity of algorithms as well as on controlled degradation of upper classification. The algorithms presented can be used with any methods of deriving a classifier where the quality of classification is monotonically decreasing function while feature set is reduced, though we have adopted the upper classifier in our study. The upper classifier has some important features that makes it suitable for database mining applications. In particular, we have shown that the upper classifier can be summarized at a desired level of abstraction by using extended decision tables. We also point out that an inconsistent decision algorithm can be interpreted as if it were a consistent decision algorithm.",
author = "Choubey, {S. K.} and Deogun, {Jitender S} and Raghavan, {V. V.} and H. Sever",
year = "1996",
month = "12",
day = "1",
language = "English (US)",
pages = "1122--1128",
note = "Proceedings of the 1996 5th IEEE International Conference on Fuzzy Systems. Part 3 (of 3) ; Conference date: 08-09-1996 Through 11-09-1996",

}

TY - CONF

T1 - Comparison of feature selection algorithms in the context of rough classifiers

AU - Choubey, S. K.

AU - Deogun, Jitender S

AU - Raghavan, V. V.

AU - Sever, H.

PY - 1996/12/1

Y1 - 1996/12/1

N2 - In this paper, we study the feature selection problem and develop and analyze four algorithms for feature selection in the context of rough set methodology. The initial state and the feasibility criterion of all these algorithms are the same, that is, they start from a given feature set and progressively remove features, while controlling the amount of degradation in classification quality, but differ in the heuristic used for pruning the search space of features. Our experimental results confirm the analytical results on the complexity of algorithms as well as on controlled degradation of upper classification. The algorithms presented can be used with any methods of deriving a classifier where the quality of classification is monotonically decreasing function while feature set is reduced, though we have adopted the upper classifier in our study. The upper classifier has some important features that makes it suitable for database mining applications. In particular, we have shown that the upper classifier can be summarized at a desired level of abstraction by using extended decision tables. We also point out that an inconsistent decision algorithm can be interpreted as if it were a consistent decision algorithm.

AB - In this paper, we study the feature selection problem and develop and analyze four algorithms for feature selection in the context of rough set methodology. The initial state and the feasibility criterion of all these algorithms are the same, that is, they start from a given feature set and progressively remove features, while controlling the amount of degradation in classification quality, but differ in the heuristic used for pruning the search space of features. Our experimental results confirm the analytical results on the complexity of algorithms as well as on controlled degradation of upper classification. The algorithms presented can be used with any methods of deriving a classifier where the quality of classification is monotonically decreasing function while feature set is reduced, though we have adopted the upper classifier in our study. The upper classifier has some important features that makes it suitable for database mining applications. In particular, we have shown that the upper classifier can be summarized at a desired level of abstraction by using extended decision tables. We also point out that an inconsistent decision algorithm can be interpreted as if it were a consistent decision algorithm.

UR - http://www.scopus.com/inward/record.url?scp=0030405336&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030405336&partnerID=8YFLogxK

M3 - Paper

SP - 1122

EP - 1128

ER -