Autonomous Knowledge-Oriented Clustering Using Decision-Theoretic Rough Set Theory

This paper processes an autonomous knowledge-oriented clustering method based on the decision-theoretic rough set theory model. In order to get the initial clustering of knowledge-oriented clusterings, the threshold values are produced autonomously in view of physics theory in this paper rather than are subjected by human intervention. Furthermore, this paper proposes a cluster validity index based on the decision-theoretic rough set theory model by considering various loss functions. Experiments with synthetic and standard data show that the novel method is not only helpful to select a termination point of the clustering algorithm, but also is useful to cluster the overlapped boundaries which is common in many data mining applications.