Improving multi-objective clustering through support vector machine: Application to gene expression data

Abstract

Microarray technology facilitates the monitoring of the expression profile of a large number of genes across different experimental conditions simultaneously. This article proposes a novel approach that combines a recently proposed multiobjective fuzzy clustering scheme with support vector machine (SVM), to yield improved solutions. The multiobjective technique is first used to produce a set of non-dominated solutions. The non-dominated set is then used to find some high-confidence points using a fuzzy voting technique. The SVM classifier is trained by this high-confidence points. Finally the remaining points are classified using the trained classifier. Results demonstrating the effectiveness of the proposed technique are provided for three real life gene expression data sets. Moreover statistical significance test has been conducted to establish the significant superiority of the proposed technique.