A hybrid approach to error reduction of support vector machines in document classification

Yoon Shik Tae; Jeong Woo Son; Mi Hwa Kong; Jun Seok Lee; Seong Bae Park; Sang Jo Lee

doi:10.1109/ITNG.2006.10

A hybrid approach to error reduction of support vector machines in document classification

Yoon Shik Tae
, Jeong Woo Son
, Mi Hwa Kong
, Jun Seok Lee
, Seong Bae Park
, Sang Jo Lee

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

1 Citation (Scopus)

Abstract

In this paper, we present a hybrid method of support vector machine and k-nearest neighbor to improve the performance of automatic text classification. The proposed methods first classifies a given document using SVM which shows the best performance in text classification, and then is reinforced by k-NN for the documents that are not confidently classified by SVM. According to the experimental results, the hybrid method achieves the F-score of 85.2, which implies that the hybrid method outperforms SVM alone.

Original language	English
Title of host publication	Proceedings - Third International Conference onInformation Technology
Subtitle of host publication	New Generations, ITNG 2006
Pages	501-506
Number of pages	6
DOIs	https://doi.org/10.1109/ITNG.2006.10
Publication status	Published - 2006
Event	Third International Conference on Information Technology: New Generations, ITNG 2006 - Las Vegas, NV, United States Duration: 10 Apr 2006 → 12 Apr 2006

Publication series

Name	Proceedings - Third International Conference onInformation Technology: New Generations, ITNG 2006
Volume	2006

Conference

Conference	Third International Conference on Information Technology: New Generations, ITNG 2006
Country/Territory	United States
City	Las Vegas, NV
Period	10/04/06 → 12/04/06

Access to Document

10.1109/ITNG.2006.10

Cite this

Tae, Y. S., Son, J. W., Kong, M. H., Lee, J. S., Park, S. B., & Lee, S. J. (2006). A hybrid approach to error reduction of support vector machines in document classification. In Proceedings - Third International Conference onInformation Technology: New Generations, ITNG 2006 (pp. 501-506). Article 1611642 (Proceedings - Third International Conference onInformation Technology: New Generations, ITNG 2006; Vol. 2006). https://doi.org/10.1109/ITNG.2006.10

@inproceedings{9a335520f3d642a682ad965d89d0eb74,

title = "A hybrid approach to error reduction of support vector machines in document classification",

abstract = "In this paper, we present a hybrid method of support vector machine and k-nearest neighbor to improve the performance of automatic text classification. The proposed methods first classifies a given document using SVM which shows the best performance in text classification, and then is reinforced by k-NN for the documents that are not confidently classified by SVM. According to the experimental results, the hybrid method achieves the F-score of 85.2, which implies that the hybrid method outperforms SVM alone.",

author = "Tae, \{Yoon Shik\} and Son, \{Jeong Woo\} and Kong, \{Mi Hwa\} and Lee, \{Jun Seok\} and Park, \{Seong Bae\} and Lee, \{Sang Jo\}",

year = "2006",

doi = "10.1109/ITNG.2006.10",

language = "English",

isbn = "0769524974",

series = "Proceedings - Third International Conference onInformation Technology: New Generations, ITNG 2006",

pages = "501--506",

booktitle = "Proceedings - Third International Conference onInformation Technology",

note = "Third International Conference on Information Technology: New Generations, ITNG 2006 ; Conference date: 10-04-2006 Through 12-04-2006",

}

Tae, YS, Son, JW, Kong, MH, Lee, JS, Park, SB & Lee, SJ 2006, A hybrid approach to error reduction of support vector machines in document classification. in Proceedings - Third International Conference onInformation Technology: New Generations, ITNG 2006., 1611642, Proceedings - Third International Conference onInformation Technology: New Generations, ITNG 2006, vol. 2006, pp. 501-506, Third International Conference on Information Technology: New Generations, ITNG 2006, Las Vegas, NV, United States, 10/04/06. https://doi.org/10.1109/ITNG.2006.10

A hybrid approach to error reduction of support vector machines in document classification. / Tae, Yoon Shik; Son, Jeong Woo; Kong, Mi Hwa et al.
Proceedings - Third International Conference onInformation Technology: New Generations, ITNG 2006. 2006. p. 501-506 1611642 (Proceedings - Third International Conference onInformation Technology: New Generations, ITNG 2006; Vol. 2006).

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

TY - GEN

T1 - A hybrid approach to error reduction of support vector machines in document classification

AU - Tae, Yoon Shik

AU - Son, Jeong Woo

AU - Kong, Mi Hwa

AU - Lee, Jun Seok

AU - Park, Seong Bae

AU - Lee, Sang Jo

PY - 2006

Y1 - 2006

N2 - In this paper, we present a hybrid method of support vector machine and k-nearest neighbor to improve the performance of automatic text classification. The proposed methods first classifies a given document using SVM which shows the best performance in text classification, and then is reinforced by k-NN for the documents that are not confidently classified by SVM. According to the experimental results, the hybrid method achieves the F-score of 85.2, which implies that the hybrid method outperforms SVM alone.

AB - In this paper, we present a hybrid method of support vector machine and k-nearest neighbor to improve the performance of automatic text classification. The proposed methods first classifies a given document using SVM which shows the best performance in text classification, and then is reinforced by k-NN for the documents that are not confidently classified by SVM. According to the experimental results, the hybrid method achieves the F-score of 85.2, which implies that the hybrid method outperforms SVM alone.

UR - https://www.scopus.com/pages/publications/33750817573

U2 - 10.1109/ITNG.2006.10

DO - 10.1109/ITNG.2006.10

M3 - Conference contribution

AN - SCOPUS:33750817573

SN - 0769524974

SN - 9780769524979

T3 - Proceedings - Third International Conference onInformation Technology: New Generations, ITNG 2006

SP - 501

EP - 506

BT - Proceedings - Third International Conference onInformation Technology

T2 - Third International Conference on Information Technology: New Generations, ITNG 2006

Y2 - 10 April 2006 through 12 April 2006

ER -

Tae YS, Son JW, Kong MH, Lee JS, Park SB, Lee SJ. A hybrid approach to error reduction of support vector machines in document classification. In Proceedings - Third International Conference onInformation Technology: New Generations, ITNG 2006. 2006. p. 501-506. 1611642. (Proceedings - Third International Conference onInformation Technology: New Generations, ITNG 2006). doi: 10.1109/ITNG.2006.10

A hybrid approach to error reduction of support vector machines in document classification

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this