Improved support vector machine generalization using normalized input space

Shawkat Ali, Kate Amanda Smith-Miles

Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review

24 Citations (Scopus)

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 languageEnglish
Title of host publicationAI 2006: Advances in Artificial Intelligence, Proceedings of 19th Australian Joint Conference on Artificial Intelligence
EditorsAlfred Hofmann
Place of PublicationGermany
PublisherSpringer
Pages362 - 371
Number of pages10
Volume4304
ISBN (Print)978-3-540-49787-5
Publication statusPublished - 2006
Externally publishedYes
EventAustralasian Joint Conference on Artificial Intelligence 2006 - Hobart, Australia
Duration: 4 Dec 20068 Dec 2006
Conference number: 19th
https://link.springer.com/book/10.1007/11941439 (Proceedings)

Conference

ConferenceAustralasian Joint Conference on Artificial Intelligence 2006
Abbreviated titleAI 2006
Country/TerritoryAustralia
CityHobart
Period4/12/068/12/06
Internet address

Cite this