Perbandingan Prestasi Pendekatan Hibrid Algoritma Genetik Dalam Pemilihan Gen-Gen Berinformatif Bagi Pengkelasan Tisu

JUMALI, RIZUAN (2007) Perbandingan Prestasi Pendekatan Hibrid Algoritma Genetik Dalam Pemilihan Gen-Gen Berinformatif Bagi Pengkelasan Tisu. Other thesis, Universiti Teknologi Malaysia.

[img]
Preview
PDF
rizuanbin_jumali.pdf

Download (2779Kb) | Preview

Abstract

With continuing researches in bioinformatics and advance technology in gene expression technology makes possible to measure and generate the expression levels of thousand of genes simultaneously compared to traditional classification methods whereby their applications are limited by existing uncertainties. According to the previous researches, there are revealed the problem using microarray data because of to select a minimal number of relevant genes whereby can maximize classification accuracy. Thus, this study used GA approach to select genes and SVM classifier for tissue classification that has better performance. The aim of this research is applying hyrid GASVM and comparing the achievement of two GA, simple GA and steady- state GA for informative genes selection from microarray gene expression data to classify diseased tissue. This comparison was performed using three types of crossover which is one-point crossover, two-point crossover and uniform crossover and also two types of selection method, roulette wheel and rank selection. Two data of leukemia cancer and colon cancer were used to be selected and then classified to evaluate the useful of the approach for small and high dimension data and also using LOOCV accuracy and testing accuracy to estimate performance of the proposed approach. Ultimately, based on the classification process, genes expression which is firstly performed by selection process on it, can effectively diagnoses diseases such as leukemia cancer and colon cancer.

Item Type: Thesis (Other)
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions: Faculty of Computer Science and Information System > Software Engineering
Depositing User: Unnamed user with email knizam@utm.my
Date Deposited: 04 Jul 2013 07:19
Last Modified: 04 Jul 2013 07:19
URI: http://ir.fsksm.utm.my/id/eprint/2736

Actions (login required)

View Item View Item