A Fast Support Vector Machine Classification Algorithm Based on Karush-Kuhn-Tucker Conditions

Ying ZHANG, Xizhao WANG, Junhai ZHAI

Research output: Book Chapters | Papers in Conference ProceedingsConference paper (refereed)Referred Conference Paperpeer-review

5 Citations (Scopus)

Abstract

Although SVM have shown potential and promising performance in classification, they have been limited by speed particularly when the training data set is large. In this paper, we propose an algorithm called the fast SVM classification algorithm based on Karush-Kuhn-Tucker (KKT) conditions. In this algorithm, we remove points that are independent of support vectors firstly in the training process, and then decompose the remaining points into blocks to accelerate the next training. From the theoretical analysis, this algorithm can remarkably reduce the computation complexity and accelerate SVM training. And experiments on both artificial and real datasets demonstrate the efficiency of this algorithm.

Original languageEnglish
Title of host publicationRough Sets, Fuzzy Sets, Data Mining and Granular Computing : 12th International Conference, RSFDGrC 2009, Proceedings
EditorsHiroshi SAKAI, Mihir Kumar CHAKRABORTY, Aboul Ella HASSANIEN, Dominik ŚLĘZAK, William ZHU
PublisherSpringer-Verlag, Berlin, Heidelberg
Pages382-389
Number of pages8
ISBN (Electronic)9783642106460
ISBN (Print)9783642106453
DOIs
Publication statusPublished - 2009
Externally publishedYes
Event12th International Conference on Rough Sets, Fuzzy Sets, Data Mining and Granular Computing, RSFDGrC 2009 - Delhi, India
Duration: 15 Dec 200918 Dec 2009

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume5908
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference12th International Conference on Rough Sets, Fuzzy Sets, Data Mining and Granular Computing, RSFDGrC 2009
Country/TerritoryIndia
CityDelhi
Period15/12/0918/12/09

Bibliographical note

This research is supported by the Natural Science Foundation of Hebei Province (F2008000635), by the key project foundation of applied fundamental research of Hebei Province (08963522D), by the plan of 100 excellent innovative scientists of the first group in Education Department of Hebei Province.

Keywords

  • Agglomerative hierarchical clustering algorithm
  • Karush-Kuhn-Tucker (KKT) conditions
  • Remove samples
  • Support Vector Machine

Fingerprint

Dive into the research topics of 'A Fast Support Vector Machine Classification Algorithm Based on Karush-Kuhn-Tucker Conditions'. Together they form a unique fingerprint.

Cite this