Effective methods for improving naive Bayes text classifiers

Sang Bum Kim; Hae-Chang Rim; Dongsuk Yook; Heui Seok Lim

Effective methods for improving naive Bayes text classifiers

Sang Bum Kim, Hae-Chang Rim, Dongsuk Yook, Heui Seok Lim

Department of Computer Science and Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

60 Citations (Scopus)

Abstract

Though naive Bayes text classifiers are widely used because of its simplicity, the techniques for improving performances of these classifiers have been rarely studied. In this paper, we propose and evaluate some general and effective techniques for improving performance of the naive Bayes text classifier. We suggest document model based parameter estimation and document length normalization to alleviate the problems in the traditional multinomial approach for text classification. In addition, Mutual-Information-weighted naive Bayes text classifier is proposed to increase the effect of highly informative words. Our techniques are evaluated on the Reuters21578 and 20 Newsgroups collections, and significant improvements are obtained over the existing multinomial naive Bayes approach.

Original language	English
Title of host publication	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Publisher	Springer Verlag
Pages	414-423
Number of pages	10
Volume	2417
ISBN (Print)	3540440380, 9783540440383
Publication status	Published - 2002
Event	7th Pacific Rim International Conference on Artificial Intelligence, PRICAI 2002 - Tokyo, Japan Duration: 2002 Aug 18 → 2002 Aug 22

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	2417
ISSN (Print)	03029743
ISSN (Electronic)	16113349

Other

Other	7th Pacific Rim International Conference on Artificial Intelligence, PRICAI 2002
Country/Territory	Japan
City	Tokyo
Period	02/8/18 → 02/8/22

ASJC Scopus subject areas

Computer Science(all)
Theoretical Computer Science

Cite this

Kim, S. B., Rim, H-C., Yook, D., & Lim, H. S. (2002). Effective methods for improving naive Bayes text classifiers. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 2417, pp. 414-423). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 2417). Springer Verlag.

Effective methods for improving naive Bayes text classifiers. / Kim, Sang Bum; Rim, Hae-Chang; Yook, Dongsuk et al.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 2417 Springer Verlag, 2002. p. 414-423 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 2417).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Kim, SB, Rim, H-C, Yook, D & Lim, HS 2002, Effective methods for improving naive Bayes text classifiers. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). vol. 2417, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 2417, Springer Verlag, pp. 414-423, 7th Pacific Rim International Conference on Artificial Intelligence, PRICAI 2002, Tokyo, Japan, 02/8/18.

Kim SB, Rim H-C, Yook D , Lim HS. Effective methods for improving naive Bayes text classifiers. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 2417. Springer Verlag. 2002. p. 414-423. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

Kim, Sang Bum ; Rim, Hae-Chang ; Yook, Dongsuk et al. / Effective methods for improving naive Bayes text classifiers. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 2417 Springer Verlag, 2002. pp. 414-423 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{db4ec0c062f44e7585c3f7ecbbcab839,

title = "Effective methods for improving naive Bayes text classifiers",

abstract = "Though naive Bayes text classifiers are widely used because of its simplicity, the techniques for improving performances of these classifiers have been rarely studied. In this paper, we propose and evaluate some general and effective techniques for improving performance of the naive Bayes text classifier. We suggest document model based parameter estimation and document length normalization to alleviate the problems in the traditional multinomial approach for text classification. In addition, Mutual-Information-weighted naive Bayes text classifier is proposed to increase the effect of highly informative words. Our techniques are evaluated on the Reuters21578 and 20 Newsgroups collections, and significant improvements are obtained over the existing multinomial naive Bayes approach.",

author = "Kim, {Sang Bum} and Hae-Chang Rim and Dongsuk Yook and Lim, {Heui Seok}",

year = "2002",

language = "English",

isbn = "3540440380",

volume = "2417",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

pages = "414--423",

booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

note = "7th Pacific Rim International Conference on Artificial Intelligence, PRICAI 2002 ; Conference date: 18-08-2002 Through 22-08-2002",

}

TY - GEN

T1 - Effective methods for improving naive Bayes text classifiers

AU - Kim, Sang Bum

AU - Rim, Hae-Chang

AU - Yook, Dongsuk

AU - Lim, Heui Seok

PY - 2002

Y1 - 2002

N2 - Though naive Bayes text classifiers are widely used because of its simplicity, the techniques for improving performances of these classifiers have been rarely studied. In this paper, we propose and evaluate some general and effective techniques for improving performance of the naive Bayes text classifier. We suggest document model based parameter estimation and document length normalization to alleviate the problems in the traditional multinomial approach for text classification. In addition, Mutual-Information-weighted naive Bayes text classifier is proposed to increase the effect of highly informative words. Our techniques are evaluated on the Reuters21578 and 20 Newsgroups collections, and significant improvements are obtained over the existing multinomial naive Bayes approach.

AB - Though naive Bayes text classifiers are widely used because of its simplicity, the techniques for improving performances of these classifiers have been rarely studied. In this paper, we propose and evaluate some general and effective techniques for improving performance of the naive Bayes text classifier. We suggest document model based parameter estimation and document length normalization to alleviate the problems in the traditional multinomial approach for text classification. In addition, Mutual-Information-weighted naive Bayes text classifier is proposed to increase the effect of highly informative words. Our techniques are evaluated on the Reuters21578 and 20 Newsgroups collections, and significant improvements are obtained over the existing multinomial naive Bayes approach.

UR - http://www.scopus.com/inward/record.url?scp=84947928349&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84947928349&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84947928349

SN - 3540440380

SN - 9783540440383

VL - 2417

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 414

EP - 423

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

PB - Springer Verlag

T2 - 7th Pacific Rim International Conference on Artificial Intelligence, PRICAI 2002

Y2 - 18 August 2002 through 22 August 2002

ER -

Effective methods for improving naive Bayes text classifiers

Abstract

Publication series

Other

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this