Multi-class Support Vector Machine (SVM) Classifiers -- An Application in Hypothyroid Detection and Classification

Chamasemani, F. F. and Singh, Y. P. (2011) Multi-class Support Vector Machine (SVM) Classifiers -- An Application in Hypothyroid Detection and Classification. In: 2011 Sixth International Conference on Bio-Inspired Computing: Theories and Applications (BIC-TA). IEEE, Bio-Inspired Computing: Theories and Applications (BIC-TA), 351 -356. ISBN 978-1-4577-1092-6

[img] Text
06046926.pdf - Published Version
Restricted to Repository staff only

Download (463kB)
Official URL: http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumb...

Abstract

The paper presents a Multi-class Support Vector Machine classifier and its application to hypothyroid detection and classification. Support Vector Machines (SVM) have been well known method in the machine learning community for binary classification problems. Multi-class SVMs (MCSVM) are usually implemented by combining several binary SVMs. The objective of this work is to show: first, robustness of various kind of kernels for Multi-class SVM classifier, second, a comparison of different constructing methods for Multi-class SVM, such as One-Against-One and One-Against-All, and finally comparing the classifiers' accuracy of Multi-class SVM classifier to AdaBoost and Decision Tree. The simulation results show that One-Against-All Support Vector Machines (OAASVM) are superior to One-Against-One Support Vector Machines (OAOSVM) with polynomial kernels. The accuracy of OAASVM is also higher than AdaBoost and Decision Tree classifier on hypothyroid disease datasets from UCI machine learning dataset.

Item Type: Book Section
Subjects: T Technology > T Technology (General)
Divisions: Faculty of Information Science and Technology (FIST)
Depositing User: Ms Suzilawati Abu Samah
Date Deposited: 01 Nov 2013 09:00
Last Modified: 01 Nov 2013 09:00
URI: http://shdl.mmu.edu.my/id/eprint/4351

Actions (login required)

View Item View Item