Title: An improved support vector machines model in medical data analysis

Authors: Hsin-Hua Chen, Ping-Feng Pai, Ying-Zhieh Cho, Fong-Chuan Lee, Ja-Chih Fu

Addresses: Division of Allergy Immunology Rheumatology, Taichung Veterans General Hospital, No. 160, Sec. 3, Chung-Kang Rd., Taichung, Taiwan. ' Department of Information Management, National Chi Nan University, 470 University Road, Puli, Nantou, Taiwan. ' Department of Information Management, National Chi Nan University, 470 University Road, Puli, Nantou, Taiwan. ' Department of Information Management, National Chi Nan University, 470 University Road, Puli, Nantou, Taiwan. ' Department of Industrial Management, National Yunlin University of Science and Technology, 123 University Road, Section 3, Douliou, Yunlin, Taiwan

Abstract: The support vector machine (SVM) technique is an emerging classification scheme that has been successfully employed in solving many classification problems. However, three main traits: features selection, dimension reduction and parameters selection, essentially influence the classification performance of SVM models. Therefore, this study developed an improved support vector machine (IMSVM) model using factor analysis (FA), kernel sliced inverse regression (KSIR) and honey-bee mating optimisation with genetic algorithms (HBMOG) to deal with feature selection, dimension reduction, and parameter selection issues, respectively, for SVM models. Then, the statlog heart dataset from the Center for Machine Learning and Intelligent Systems at the University of California, Irvine (UCI) was used to demonstrate the performance of the IMSVM model. Experimental results revealed that the IMSVM model can provide more accurate classification results than the results obtained by classification models in previous literature. Thus, the proposed model is a promising alternative for analysing medical data.

Keywords: support vector machines; SVM; factor analysis; honey bee mating optimisation; genetic algorithms; kernel sliced inverse regression; medical data analysis; classification; feature selection; dimension reduction; parameter selection.

DOI: 10.1504/IJMMNO.2010.031747

International Journal of Mathematical Modelling and Numerical Optimisation, 2010 Vol.1 No.3, pp.168 - 184

Published online: 22 Feb 2010 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article