Abstract:
The relentless development of Microarray technology have meant that the dimen-
sionality of data that is produced by the Microarray chips have increased many
folds over the years. Recognition of patterns and other subsequent analysis from
the thousands of gene expression values is particularly di cult and primary role of
an e ective feature selection is to simplify this task. Removal of less informative
genes helps to alleviate the e ects of noise and redundancy, and simpli es the task
of disease classi cation and prediction of medical conditions such as cancer. In
this study the shortcoming of the current GA based approach for feature selection
has been improved. A lter and wrapper models are put to use to take advantage
of the facilities that each provides. As lter method exhibits some limitations, in
this study an approach to ltering (BW ratio) has been employed. As a wrapper
approach Particle Swarm Optimization (PSO) has been proposed.
Description:
Supervised by
Prof. Dr. M.A. Mottalib
Head of the Department,
Department of Computer Science and Engineering (CSE),
Islamic University of Technology (IUT).
Shaikh Jeeshan Kabeer,
Co-supervisor,
Lecturer,
Department of Computer Science and Engineering (CSE),
Islamic University of Technology(IUT).
