A Bat K-Medoids Clustering Algorithm with Clonal Selection Technique for DNA Microarray Data

In bioinformatics fields, researchers analyze tumors at the molecular level using DNA microarray data, and further classify different tumor subtypes according to the different expressions of related pathogenic genes in the same tumor. High dimensions and small samples is characteristic of DNA microarray data sets. In this paper, we proposed an improved bat clonal K-medoids clustering algorithm (IBACK-medoids). In the IBACK-medoids, we improved standard bat algorithm through accelerating the convergence speed. Furthermore, we introduce the main idea of clonal selection algorithm in the improved bat algorithm. Finally, to obtain more better clustering convergence, the improved bat clonal algorithm and K-medoids algorithm are combined. In experiments, the evaluation tools for clustering algorithms are F -measure, Entropy and Purity. The experimental results show the effectiveness of our proposed algorithms.