The New Jersey Institute of Technology's
Electronic Theses & Dissertations Project

Title:	SLIMSVM : a simple implementation of support vector machine for analysis of microarray data
Author:	Karmaker, Avik
View Online:	njit-etd2004-113 (xiii, 81 pages ~ 4.0 MB pdf)
Department:	College of Computing Sciences
Degree:	Master of Science
Program:	Computational Biology
Document Type:	Thesis
Advisory Committee:	Ma, Qun (Committee chair) Roshan, Usman W. (Committee member) Shih, Frank Y. (Committee member)
Date:	2004-08
Keywords:	Machine learning techniques Microarray data analysis
Availability:	Unrestricted
Abstract:	Support Vector Machine (SVM) is a supervised machine learning technique being widely used in multiple areas of biological analysis including microarray data analysis. SlimSVM has been developed with the intention of replacing OSU SVM as the classification component of GenoIterSVM in order to make it independent of other SVM packages. GenolterSVM, developed by Dr. Marc Ma, is a SVM implementation with an iterative refinement algorithm for improved accuracy of classification of genotype microarray data. SlimSVM is an object-oriented, modular, and easy-to-use implementation written in C++. It supports dot (linear) and polynomial (non-linear) kernels. The program has been tested with artificial non-biological and microarray data. Testing with microarray data was performed to observe how SlimSVM handles medium-sized data files (containing thousands of data points) since it would ultimately be used to analyze them. The results were compared to those of LIBSVM, a leading SVM software, and the comparison demonstrates that implementation of SlimS VM was carried out accurately.