Improved support vector machine generalization using normalized input space

Shawkat Ali; Kate Amanda Smith-Miles

Improved support vector machine generalization using normalized input space

Shawkat Ali, Kate Amanda Smith-Miles

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

Abstract

Data pre-processing always plays a key role in learning algorithm performance. In this research we consider data pre-processing by normalization for Support Vector Machines (SVMs). We examine the normalization affect across 112 classification problems with SVM using the rbf kernel. We observe a significant classification improvement due to normalization. Finally we suggest a rule based method to find when normalization is necessary for a specific classification problem. The best normalization method is also automatically selected by SVM itself.

Original language	English
Title of host publication	AI 2006: Advances in Artificial Intelligence, Proceedings of 19th Australian Joint Conference on Artificial Intelligence
Editors	Alfred Hofmann
Place of Publication	Germany
Publisher	Springer
Pages	362 - 371
Number of pages	10
Volume	4304
ISBN (Print)	978-3-540-49787-5
Publication status	Published - 2006
Externally published	Yes
Event	Australasian Joint Conference on Artificial Intelligence 2006 - Hobart, Australia Duration: 4 Dec 2006 → 8 Dec 2006 Conference number: 19th https://link.springer.com/book/10.1007/11941439 (Proceedings)

Conference

Conference	Australasian Joint Conference on Artificial Intelligence 2006
Abbreviated title	AI 2006
Country/Territory	Australia
City	Hobart
Period	4/12/06 → 8/12/06
Internet address	https://link.springer.com/book/10.1007/11941439 (Proceedings)

Cite this

@inproceedings{0d2191cf8e974e25b0cc695644acfa91,

title = "Improved support vector machine generalization using normalized input space",

abstract = "Data pre-processing always plays a key role in learning algorithm performance. In this research we consider data pre-processing by normalization for Support Vector Machines (SVMs). We examine the normalization affect across 112 classification problems with SVM using the rbf kernel. We observe a significant classification improvement due to normalization. Finally we suggest a rule based method to find when normalization is necessary for a specific classification problem. The best normalization method is also automatically selected by SVM itself.",

author = "Shawkat Ali and Smith-Miles, \{Kate Amanda\}",

year = "2006",

language = "English",

isbn = "978-3-540-49787-5",

volume = "4304",

pages = "362 -- 371",

editor = "Alfred Hofmann",

booktitle = "AI 2006: Advances in Artificial Intelligence, Proceedings of 19th Australian Joint Conference on Artificial Intelligence",

publisher = "Springer",

address = "Switzerland",

note = "Australasian Joint Conference on Artificial Intelligence 2006, AI 2006 ; Conference date: 04-12-2006 Through 08-12-2006",

url = "https://link.springer.com/book/10.1007/11941439",

}

Ali, S & Smith-Miles, KA 2006, Improved support vector machine generalization using normalized input space. in A Hofmann (ed.), AI 2006: Advances in Artificial Intelligence, Proceedings of 19th Australian Joint Conference on Artificial Intelligence. vol. 4304, Springer, Germany, pp. 362 - 371, Australasian Joint Conference on Artificial Intelligence 2006, Hobart, Tasmania, Australia, 4/12/06.

Improved support vector machine generalization using normalized input space. / Ali, Shawkat; Smith-Miles, Kate Amanda.
AI 2006: Advances in Artificial Intelligence, Proceedings of 19th Australian Joint Conference on Artificial Intelligence. ed. / Alfred Hofmann. Vol. 4304 Germany: Springer, 2006. p. 362 - 371.

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

TY - GEN

T1 - Improved support vector machine generalization using normalized input space

AU - Ali, Shawkat

AU - Smith-Miles, Kate Amanda

N1 - Conference code: 19th

PY - 2006

Y1 - 2006

N2 - Data pre-processing always plays a key role in learning algorithm performance. In this research we consider data pre-processing by normalization for Support Vector Machines (SVMs). We examine the normalization affect across 112 classification problems with SVM using the rbf kernel. We observe a significant classification improvement due to normalization. Finally we suggest a rule based method to find when normalization is necessary for a specific classification problem. The best normalization method is also automatically selected by SVM itself.

AB - Data pre-processing always plays a key role in learning algorithm performance. In this research we consider data pre-processing by normalization for Support Vector Machines (SVMs). We examine the normalization affect across 112 classification problems with SVM using the rbf kernel. We observe a significant classification improvement due to normalization. Finally we suggest a rule based method to find when normalization is necessary for a specific classification problem. The best normalization method is also automatically selected by SVM itself.

UR - http://www.springerlink.com/content/241p276441411p04/

UR - https://www.scopus.com/pages/publications/77953160967

M3 - Conference Paper

SN - 978-3-540-49787-5

VL - 4304

SP - 362

EP - 371

BT - AI 2006: Advances in Artificial Intelligence, Proceedings of 19th Australian Joint Conference on Artificial Intelligence

A2 - Hofmann, Alfred

PB - Springer

CY - Germany

T2 - Australasian Joint Conference on Artificial Intelligence 2006

Y2 - 4 December 2006 through 8 December 2006

ER -

Improved support vector machine generalization using normalized input space

Abstract

Conference

Other files and links

Cite this