A Hybrid Approach to Error Reduction of Support Vector Machines in Document Classification

I
ITNG
2006
Third International Conference on Information Technology: New Generations (ITNG'06)
Abstract - A Hybrid Approach to Error Reduction of Support Vector Machines in Document Classification

	This Article
	Subscribers, please Login Purchase article: $19 PDF IEEE Xplore Subscribers RSS feed
	Share
	Email this Article to a friend
	Bibliographic References
	ASCII Text BibTex RefWorks Procite/RefMan
	Add to:
	Digg Furl Spurl Blink Simpy Google Del.icio.us Y!MyWeb
	Search
	Similar Articles Articles by Yoon-Shik Tae Articles by Jeong woo Son Articles by Mi-hwa Kong Articles by Jun-Seok Lee Articles by Seong-Bae Park Articles by Sang-Jo Lee

Third International Conference on Information Technology: New Generations (ITNG'06)

Las Vegas, Nevada

April 10-April 12

ISBN: 0-7695-2497-4

	ASCII Text	x
	Yoon-Shik Tae, Jeong woo Son, Mi-hwa Kong, Jun-Seok Lee, Seong-Bae Park, Sang-Jo Lee, "A Hybrid Approach to Error Reduction of Support Vector Machines in Document Classification," Information Technology: New Generations, Third International Conference on, pp. 501-506, Third International Conference on Information Technology: New Generations (ITNG'06), 2006.

	BibTex	x
	@article{ 10.1109/ITNG.2006.10, author = {Yoon-Shik Tae and Jeong woo Son and Mi-hwa Kong and Jun-Seok Lee and Seong-Bae Park and Sang-Jo Lee}, title = {A Hybrid Approach to Error Reduction of Support Vector Machines in Document Classification}, journal ={Information Technology: New Generations, Third International Conference on}, volume = {0}, year = {2006}, isbn = {0-7695-2497-4}, pages = {501-506}, doi = {http://doi.ieeecomputersociety.org/10.1109/ITNG.2006.10}, publisher = {IEEE Computer Society}, address = {Los Alamitos, CA, USA}, }

	RefWorks Procite/RefMan/Endnote	x
	TY - CONF JO - Information Technology: New Generations, Third International Conference on TI - A Hybrid Approach to Error Reduction of Support Vector Machines in Document Classification SN - 0-7695-2497-4 SP501 EP506 A1 - Yoon-Shik Tae, A1 - Jeong woo Son, A1 - Mi-hwa Kong, A1 - Jun-Seok Lee, A1 - Seong-Bae Park, A1 - Sang-Jo Lee, PY - 2006 KW - null VL - 0 JA - Information Technology: New Generations, Third International Conference on ER -

Yoon-Shik Tae, Kyungpook National University, Korea

Jeong woo Son, Kyungpook National University, Korea

Mi-hwa Kong, Kyungpook National University, Korea

Jun-Seok Lee, Kyungpook National University, Korea

Seong-Bae Park, Kyungpook National University, Korea

Sang-Jo Lee, Kyungpook National University, Korea

DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/ITNG.2006.10

In this paper, we present a hybrid method fo support vector machine and k-nearest neighbor to improve the performance of automatic text classifcation. The proposed methods first classifies a given document using SVM which shows the best performance in text classification, and then is reinforcd 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 tha the hybrid method outperforms SVM alone.

Citation:

Yoon-Shik Tae, Jeong woo Son, Mi-hwa Kong, Jun-Seok Lee, Seong-Bae Park, Sang-Jo Lee, "A Hybrid Approach to Error Reduction of Support Vector Machines in Document Classification," itng, pp.501-506, Third International Conference on Information Technology: New Generations (ITNG'06), 2006

Peer Review Notice | Give Us Feedback

Usage of this product signifies your acceptance of the Terms of Use.