A semiautonomous clustering algorithm based on decision-theoretic rough set theory

The clusters tend to have vague or imprecise boundaries in some fields such as web mining, since clustering has been widely used. Decision-theoretic rough set model (DTRSM) is a typical probabilistic rough set model, which has the ability to deal with imprecise, uncertain, and vague information. Therefore, a novel clustering algorithm based on the DTRSM is proposed in this paper, which can decide the overlapping boundary through a loss function given by users. Furthermore, in order to determine the initial clustering autonomously, a threshold values computing method, select differences, is developed based on the knowledge-oriented clustering framework. The select differences method reduces the time complexity of computing the initial threshold values to O(nlgn). The experimental results show that the new algorithm is valid and efficient.