Smart: Novel self splitting-merging clustering algorithm

In this paper, we propose a new self splitting-merging clustering algorithm, named splitting-merging awareness tactics (SMART). The novel framework, which integrates many techniques, starts with one cluster and employs a splitting-while-merging process. The SMART has self-awareness to split and merge the clusters automatically in iterations. Both the framework and the techniques are detailed and illustrated by a good benchmark test. Furthermore, three microarray gene expression datasets are studied using our approach. The numerical results show that our proposal is automotive and effective.