Model Based Modified K-Means Clustering for Microarray Data

Large amount of gene expression data obtained from microarray technologies should be analyzed and interpreted in appropriate manner for the benefit of researchers. Using microarray techniques one can monitor the expressions levels of thousands of genes simultaneously. One challenging problem in gene expression analysis is to define the number of clusters. This can be done by some efficient clustering techniques; the model based modified k-means method introduced in this paper could find the exact number of clusters and overcome the problems in the existing k-means clustering technique. Our experimental results show the efficiency of our method by calculating and comparing the sum of squares with different k values.