Efficient multi-way text categorization via generalized discriminant analysis

Tao Li, Shenghuo Zhu, Mitsunori Ogihara

Research output: Chapter in Book/Report/Conference proceedingConference contribution

20 Citations (Scopus)

Abstract

Text categorization is an important research area and has been receiving much attention due to the growth of the on-line information and of Internet. Automated text categorization is generally cast as a multi-class classification problem. Much of previous work focused on binary document classification problems. Support vector machines (SVMs) excel in binary classification, but the elegant theory behind large-margin hyperplane cannot be easily extended to multi-class text classification. In addition, the training time and scaling are also important concerns. On the other hand, other techniques naturally extensible to handle multi-class classification are generally not as accurate as SVM. This paper presents a simple and efficient solution to multi-class text categorization. Classification problems are first formulated as optimization via discriminant analysis. Text categorization is then cast as the problem of finding coordinate transformations that reflects the inherent similarity from the data. While most of the previous approaches decompose a multi-class classification problem into multiple independent binary classification tasks, the proposed approach enables direct multi-class classification. By using Generalized Singular Value Decomposition (GSVD), a coordinate transformation that reflects the inherent class structure indicated by the generalized singular values is identified. Extensive experiments demonstrate the efficiency and effectiveness of the proposed approach.

Original languageEnglish (US)
Title of host publicationInternational Conference on Information and Knowledge Management, Proceedings
EditorsO. Frieder, J. Hammer, S. Qureshi, L. Seligman
Pages317-324
Number of pages8
StatePublished - 2003
Externally publishedYes
EventCIKM 2003: Proceedings of the Twelfth ACM International Conference on Information and Knowledge Management - New Orleans, LA, United States
Duration: Nov 3 2003Nov 8 2003

Other

OtherCIKM 2003: Proceedings of the Twelfth ACM International Conference on Information and Knowledge Management
CountryUnited States
CityNew Orleans, LA
Period11/3/0311/8/03

Fingerprint

Text categorization
Discriminant analysis
Support vector machine
Scaling
Text classification
Document classification
World Wide Web
Experiment
Excel
Singular value decomposition
Margin

Keywords

  • Analysis
  • Discriminant
  • GSVD
  • Multi-class Text Categorization

ASJC Scopus subject areas

  • Business, Management and Accounting(all)

Cite this

Li, T., Zhu, S., & Ogihara, M. (2003). Efficient multi-way text categorization via generalized discriminant analysis. In O. Frieder, J. Hammer, S. Qureshi, & L. Seligman (Eds.), International Conference on Information and Knowledge Management, Proceedings (pp. 317-324)

Efficient multi-way text categorization via generalized discriminant analysis. / Li, Tao; Zhu, Shenghuo; Ogihara, Mitsunori.

International Conference on Information and Knowledge Management, Proceedings. ed. / O. Frieder; J. Hammer; S. Qureshi; L. Seligman. 2003. p. 317-324.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Li, T, Zhu, S & Ogihara, M 2003, Efficient multi-way text categorization via generalized discriminant analysis. in O Frieder, J Hammer, S Qureshi & L Seligman (eds), International Conference on Information and Knowledge Management, Proceedings. pp. 317-324, CIKM 2003: Proceedings of the Twelfth ACM International Conference on Information and Knowledge Management, New Orleans, LA, United States, 11/3/03.
Li T, Zhu S, Ogihara M. Efficient multi-way text categorization via generalized discriminant analysis. In Frieder O, Hammer J, Qureshi S, Seligman L, editors, International Conference on Information and Knowledge Management, Proceedings. 2003. p. 317-324
Li, Tao ; Zhu, Shenghuo ; Ogihara, Mitsunori. / Efficient multi-way text categorization via generalized discriminant analysis. International Conference on Information and Knowledge Management, Proceedings. editor / O. Frieder ; J. Hammer ; S. Qureshi ; L. Seligman. 2003. pp. 317-324
@inproceedings{c8d7dc1410c64b0b87462017c0fcb56e,
title = "Efficient multi-way text categorization via generalized discriminant analysis",
abstract = "Text categorization is an important research area and has been receiving much attention due to the growth of the on-line information and of Internet. Automated text categorization is generally cast as a multi-class classification problem. Much of previous work focused on binary document classification problems. Support vector machines (SVMs) excel in binary classification, but the elegant theory behind large-margin hyperplane cannot be easily extended to multi-class text classification. In addition, the training time and scaling are also important concerns. On the other hand, other techniques naturally extensible to handle multi-class classification are generally not as accurate as SVM. This paper presents a simple and efficient solution to multi-class text categorization. Classification problems are first formulated as optimization via discriminant analysis. Text categorization is then cast as the problem of finding coordinate transformations that reflects the inherent similarity from the data. While most of the previous approaches decompose a multi-class classification problem into multiple independent binary classification tasks, the proposed approach enables direct multi-class classification. By using Generalized Singular Value Decomposition (GSVD), a coordinate transformation that reflects the inherent class structure indicated by the generalized singular values is identified. Extensive experiments demonstrate the efficiency and effectiveness of the proposed approach.",
keywords = "Analysis, Discriminant, GSVD, Multi-class Text Categorization",
author = "Tao Li and Shenghuo Zhu and Mitsunori Ogihara",
year = "2003",
language = "English (US)",
pages = "317--324",
editor = "O. Frieder and J. Hammer and S. Qureshi and L. Seligman",
booktitle = "International Conference on Information and Knowledge Management, Proceedings",

}

TY - GEN

T1 - Efficient multi-way text categorization via generalized discriminant analysis

AU - Li, Tao

AU - Zhu, Shenghuo

AU - Ogihara, Mitsunori

PY - 2003

Y1 - 2003

N2 - Text categorization is an important research area and has been receiving much attention due to the growth of the on-line information and of Internet. Automated text categorization is generally cast as a multi-class classification problem. Much of previous work focused on binary document classification problems. Support vector machines (SVMs) excel in binary classification, but the elegant theory behind large-margin hyperplane cannot be easily extended to multi-class text classification. In addition, the training time and scaling are also important concerns. On the other hand, other techniques naturally extensible to handle multi-class classification are generally not as accurate as SVM. This paper presents a simple and efficient solution to multi-class text categorization. Classification problems are first formulated as optimization via discriminant analysis. Text categorization is then cast as the problem of finding coordinate transformations that reflects the inherent similarity from the data. While most of the previous approaches decompose a multi-class classification problem into multiple independent binary classification tasks, the proposed approach enables direct multi-class classification. By using Generalized Singular Value Decomposition (GSVD), a coordinate transformation that reflects the inherent class structure indicated by the generalized singular values is identified. Extensive experiments demonstrate the efficiency and effectiveness of the proposed approach.

AB - Text categorization is an important research area and has been receiving much attention due to the growth of the on-line information and of Internet. Automated text categorization is generally cast as a multi-class classification problem. Much of previous work focused on binary document classification problems. Support vector machines (SVMs) excel in binary classification, but the elegant theory behind large-margin hyperplane cannot be easily extended to multi-class text classification. In addition, the training time and scaling are also important concerns. On the other hand, other techniques naturally extensible to handle multi-class classification are generally not as accurate as SVM. This paper presents a simple and efficient solution to multi-class text categorization. Classification problems are first formulated as optimization via discriminant analysis. Text categorization is then cast as the problem of finding coordinate transformations that reflects the inherent similarity from the data. While most of the previous approaches decompose a multi-class classification problem into multiple independent binary classification tasks, the proposed approach enables direct multi-class classification. By using Generalized Singular Value Decomposition (GSVD), a coordinate transformation that reflects the inherent class structure indicated by the generalized singular values is identified. Extensive experiments demonstrate the efficiency and effectiveness of the proposed approach.

KW - Analysis

KW - Discriminant

KW - GSVD

KW - Multi-class Text Categorization

UR - http://www.scopus.com/inward/record.url?scp=7244220821&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=7244220821&partnerID=8YFLogxK

M3 - Conference contribution

SP - 317

EP - 324

BT - International Conference on Information and Knowledge Management, Proceedings

A2 - Frieder, O.

A2 - Hammer, J.

A2 - Qureshi, S.

A2 - Seligman, L.

ER -