Novel Extension of k − TSP Algorithm for Microarray Classification
This paper presents a new method, referred as Weight k − TSP, which generates simple and accurate decision rules that can be widely used for classifying gene expression data. The proposed method extends previous approaches: TSP and k − TSP algorithms by considering weight pairwise mRNA comparisons and percentage changes of gene expressions in different classes. Both rankings have been modified as well as decision rules, however the concept of ”relative expression reversals” is retained. New solutions to match analyzed datasets more accurately were also included. Experimental validation was performed on several human microarray datasets and obtained results are promising.
KeywordsSupport Vector Machine Microarray Dataset Fuzzy Support Vector Machine Simple Decision Rule Secondary Ranking
Unable to display preview. Download preview PDF.
- 3.Cho, H.S., Kim, T.S., Wee, J.W.: cDNA Microarray Data Based Classification of Cancers Using Neural Networks and Genetic Algorithms. Nanotech 1 (2003)Google Scholar
- 8.Grześ, M., Krȩtowski, M.: Decision Tree Approach to Microarray Data Analysis. Biocybernetics and Biomedical Engineering 27(3), 29–42 (2007)Google Scholar
- 10.Kononenko, I.: Estimating Attributes: Analysis and Extensions of RELIEF. In: European Conference on Machine Learning, pp. 171–182 (1994)Google Scholar
- 13.Mao, Y., Zhou, X., Pi, D., Sun, Y., Wong, S.T.C.: Multiclass Cancer Classification by Using Fuzzy Support Vector Machine and Binary Decision Tree With Gene Selection. J. Biomed. Biotechnol., 160–171 (2005)Google Scholar
- 22.Zhang, C., Li, P., Rajendran, A., Deng, Y., Chen, D.: Parallelization of multicategory support vector machines (PMC-SVM) from classifying microarray data. BMC Bioinformatics (2006)Google Scholar