Review of Gene Subset Selection using Modified K-Nearest Neighbor Clustering Algorithm

In analysis of microarray data using subset selection are mainly used for classification and prediction of genes and evaluate the discern ability of gene for specific diseases. The Performance of Modified K-Nearest Neighbor (MKNN) technique is follows as the robustness function for Genetic Algorithm (GA) based on the classification principle. In this work, MKNN is tested with various dataset for different probability functions such T-value, Score, BW and GA/KNN. This work proposed a method modified K-nearest neighbor algorithm using probability mechanism to improve the performance of gene selection by means of distance measure and accuracy. Proposed method provides better results and highly suitable for effective feature selection.