Tintu Thomas
C U Shah Medical College
India
Title: ANFIS based approach for mining high dimensional gene expression microarray data: A comparative study
Biography
Biography: Tintu Thomas
Abstract
Microarrays technology is used to find the expression of many thousands of genes simultaneously. Identification and classification of set of genes out of these thousands is a complex process. In microarray data, a major challenge is the need of robust method for proper identification of differential expression genes. In this paper, we made a comparative study on the genomic classification performance of conventional neural networks method and fuzzy inference method. We used fuzzy inference based classification rules for extracting significant genes from the gene expression data set. Fuzzy rules were utilized to train the Fuzzy Inference System (FIS) and classified the gene expression level in to useful output form namely expressed and non-expressed genes. It is found that the adaptive neural network fuzzy inference methods worked better in classifying differentially expressed genes compared to other conventional methods.