IDENTIFYING POTENTIAL GENE MARKERS USING SVM CLASSIFIER ENSEMBLE
An important task in modern data mining is to utilize advanced data analysis and integration tools in gene expression pattern discovery and classification. These tools include a number of machine learning techniques, which may help in identifying relevant features for diagnostic and system biology studies. Furthermore, discovery of novel automated techniques for intelligent information retrieval and knowledge representation are crucial for biological data analysis. When a living cell undergoes a biological process, not all of its genes are expressed at the same time. Function of a cell is critically related to the gene expression at a given time and their relative abundance. For understanding biological processes, it is usual to measure gene expression levels in different developmental phases, body tissues, clinical conditions, and organisms. This information of differential gene expression can be utilized in characterizing gene function, determining experimental treatment effects, and understanding other molecular biological processes. Traditional approaches to genomic research was based on examining and collecting data for a single gene locally. The progress in the field of microarray technology has made possible to the study of the expression levels of a large number of genes across different time points or tumor samples ...