Efficiently mining gene expression data via integrated clustering and validation techniques

Abstract

In recent years, the microarray techniques have received extensive attentions due to its wide applications in biomedical industry. The main advantage of microarray technique is it allows simultaneous studies of the expressions of thousands of genes in a single experiment. Analyzing the microarray data is a challenge that arises the applications of various clustering methods used for data mining. Although a number of clustering methods have been proposed, they can not meet the requirements of automation, high quality and high efficiency at the same time in analyzing gene expression data. In this paper, we propose an automatic and efficient clustering approach for mining gene expression data produced via microarray techniques. Through performance experiments on real data sets, the proposed method is shown to achieve higher efficiency, clustering quality and automation than other clustering methods.