An SMO Approach to fast SVM for classification of large scale data

Juanxi Lin; Mengnan Song; Takayuki Furuzuki

doi:10.1109/ICITCS.2014.7021735

An SMO Approach to fast SVM for classification of large scale data

Juanxi Lin, Mengnan Song, Takayuki Furuzuki

Graduate School of Information, Production and Systems

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

Abstract

In this paper, a novel approach is proposed as a new fast Support Vector Machines (SVM) basing on sequential minimal optimization(SMO), minimum enclosing ball(MEB) approach and active set strategy. The combination with these 3 techniques largely accelerates the training process of SVM, attains fewer support vectors(SVs) as well as obtains a acceptable accuracy comparing to original SVM. From simulation results, it is stated that the proposed method will be a good alternative for classification of large scale data.

Original language	English
Title of host publication	2014 International Conference on IT Convergence and Security, ICITCS 2014
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781479965410
DOIs	https://doi.org/10.1109/ICITCS.2014.7021735
Publication status	Published - 2014 Jan 23
Event	4th 2014 International Conference on IT Convergence and Security, ICITCS 2014 - Beijing, China Duration: 2014 Oct 28 → 2014 Oct 30

Other

Other	4th 2014 International Conference on IT Convergence and Security, ICITCS 2014
Country	China
City	Beijing
Period	14/10/28 → 14/10/30

Keywords

active set
minimum enclosing ball
sequential minimal optimization
Support Vector Machine

ASJC Scopus subject areas

Computer Networks and Communications

Access to Document

10.1109/ICITCS.2014.7021735

Fingerprint Dive into the research topics of 'An SMO Approach to fast SVM for classification of large scale data'. Together they form a unique fingerprint.

Cite this

Lin, J, Song, M & Furuzuki, T 2014, An SMO Approach to fast SVM for classification of large scale data. in 2014 International Conference on IT Convergence and Security, ICITCS 2014., 7021735, Institute of Electrical and Electronics Engineers Inc., 4th 2014 International Conference on IT Convergence and Security, ICITCS 2014, Beijing, China, 14/10/28. https://doi.org/10.1109/ICITCS.2014.7021735

@inproceedings{28fe094c8f54493a8dd9ce146773e892,

title = "An SMO Approach to fast SVM for classification of large scale data",

abstract = "In this paper, a novel approach is proposed as a new fast Support Vector Machines (SVM) basing on sequential minimal optimization(SMO), minimum enclosing ball(MEB) approach and active set strategy. The combination with these 3 techniques largely accelerates the training process of SVM, attains fewer support vectors(SVs) as well as obtains a acceptable accuracy comparing to original SVM. From simulation results, it is stated that the proposed method will be a good alternative for classification of large scale data.",

keywords = "active set, minimum enclosing ball, sequential minimal optimization, Support Vector Machine",

author = "Juanxi Lin and Mengnan Song and Takayuki Furuzuki",

year = "2014",

month = jan,

day = "23",

doi = "10.1109/ICITCS.2014.7021735",

language = "English",

booktitle = "2014 International Conference on IT Convergence and Security, ICITCS 2014",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

address = "United States",

note = "4th 2014 International Conference on IT Convergence and Security, ICITCS 2014 ; Conference date: 28-10-2014 Through 30-10-2014",

}

TY - GEN

T1 - An SMO Approach to fast SVM for classification of large scale data

AU - Lin, Juanxi

AU - Song, Mengnan

AU - Furuzuki, Takayuki

PY - 2014/1/23

Y1 - 2014/1/23

N2 - In this paper, a novel approach is proposed as a new fast Support Vector Machines (SVM) basing on sequential minimal optimization(SMO), minimum enclosing ball(MEB) approach and active set strategy. The combination with these 3 techniques largely accelerates the training process of SVM, attains fewer support vectors(SVs) as well as obtains a acceptable accuracy comparing to original SVM. From simulation results, it is stated that the proposed method will be a good alternative for classification of large scale data.

AB - In this paper, a novel approach is proposed as a new fast Support Vector Machines (SVM) basing on sequential minimal optimization(SMO), minimum enclosing ball(MEB) approach and active set strategy. The combination with these 3 techniques largely accelerates the training process of SVM, attains fewer support vectors(SVs) as well as obtains a acceptable accuracy comparing to original SVM. From simulation results, it is stated that the proposed method will be a good alternative for classification of large scale data.

KW - active set

KW - minimum enclosing ball

KW - sequential minimal optimization

KW - Support Vector Machine

UR - http://www.scopus.com/inward/record.url?scp=84988259230&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84988259230&partnerID=8YFLogxK

U2 - 10.1109/ICITCS.2014.7021735

DO - 10.1109/ICITCS.2014.7021735

M3 - Conference contribution

AN - SCOPUS:84988259230

BT - 2014 International Conference on IT Convergence and Security, ICITCS 2014

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 4th 2014 International Conference on IT Convergence and Security, ICITCS 2014

Y2 - 28 October 2014 through 30 October 2014

ER -