A Hybrid PSO-FSVM Model and Its Application to Imbalanced Classification of Mammograms

Samma, Hussein, Lim, Chee Peng and Ngah, Umi Kalthum 2013, A Hybrid PSO-FSVM Model and Its Application to Imbalanced Classification of Mammograms, in Intelligent information and database systems: 5th Asian conference, ACIIDS 2013, Kuala Lumpur, Malaysia, March 18-20, 2013, proceedings : Part 1, Springer, Berlin, Germany, pp.275-284.

Attached Files
Name Description MIMEType Size Downloads

Title A Hybrid PSO-FSVM Model and Its Application to Imbalanced Classification of Mammograms
Author(s) Samma, Hussein
Lim, Chee Peng
Ngah, Umi Kalthum
Title of book Intelligent information and database systems: 5th Asian conference, ACIIDS 2013, Kuala Lumpur, Malaysia, March 18-20, 2013, proceedings : Part 1
Editor(s) Selamat, Ali
Nguyen, Ngoc Thanh
Haron, Habibollah
Publication date 2013
Chapter number 29
Total chapters 52
Start page 275
End page 284
Total pages 10
Publisher Springer
Place of Publication Berlin, Germany
Summary In this work, a hybrid model comprising Particle Swarm Optimization (PSO) and the Fuzzy Support Vector Machine (FSVM) for tackling imbalanced classification problems is proposed. A PSO algorithm, guided by the G-mean measure, is used to optimize the FSVM parameters in imbalanced classification problems. The hybrid PSO-FSVM model is evaluated using a mammogram mass classification problem. The experimental results are analyzed and compared with those from other methods. The outcomes positively demonstrate that the proposed PSO-FSVM model is able to achieve comparable, if not better, results for imbalanced data classification problems.
ISBN 9783642365454
9783642365461
Language eng
Field of Research 080108 Neural, Evolutionary and Fuzzy Computation
Socio Economic Objective 920203 Diagnostic Methods
HERDC Research category B1 Book chapter
Copyright notice ©2013, Springer
Persistent URL http://hdl.handle.net/10536/DRO/DU:30057218

Document type: Book Chapter
Collection: Centre for Intelligent Systems Research
Connect to link resolver
 
Unless expressly stated otherwise, the copyright for items in DRO is owned by the author, with all rights reserved.

Versions
Version Filter Type
Access Statistics: 27 Abstract Views, 9 File Downloads  -  Detailed Statistics
Created: Wed, 23 Oct 2013, 11:12:22 EST

Every reasonable effort has been made to ensure that permission has been obtained for items included in DRO. If you believe that your rights have been infringed by this repository, please contact drosupport@deakin.edu.au.