A merging-based consensus-driven fuzzy clustering of distributed data

With consensus tendency, fuzzy clustering results are given in advance and the goal is to reconcile the results by building aggregate partition and prototypes. In this paper, we present a consensus-driven fuzzy clustering algorithm in which distributed data are progressively merged into intersites´ clusters through global aggregation based on similarity relationships. The ensuring algorithm considers dispersion and dissimilarity between local partitions while respecting the confidentiality constraints which prohibits the data sharing. Several illustrative numeric tests using both synthetic data and those coming from publicity available machine learning repositories are also included. The experimental component of the study shows the efficiency of the proposed algorithm.