Quality-Driven Hierarchical Clustering Algorithm for Service Intelligence Computation

摘要: Clustering is an important technique for intelligence computation such as trust, recommendation, reputation, and requirement elicitation. With the user centric nature of service user's lack prior knowledge on distribution raw data, one challenge how to associate quality requirements clustering results with algorithmic output properties (e.g. number clusters be targeted). In this paper, we focus hierarchical process propose two quality-driven algorithms, HBH (homogeneity-based hierarchical) HDH (homogeneity-driven minimum acceptable homogeneity relative population each cluster their input criteria. Furthermore, also give a HDH-approximation algorithm in order address time performance issue. Experimental study data sets different density dispersion levels shows that gives best result can significantly improve execution time.