PROJECT TITLE :

Uncertain Data Clustering in Distributed Peer-to-Peer Networks - 2017

ABSTRACT:

Uncertain data clustering has been recognized as a vital task in the research of information mining. Several centralized clustering algorithms are extended by defining new distance or similarity measurements to tackle this issue. With the fast development of network applications, these centralized methods show their limitations in conducting data clustering during a large dynamic distributed peer-to-peer network thanks to the privacy and security issues or the technical constraints brought by distributive environments. In this paper, we tend to propose a novel distributed unsure data clustering algorithm, in that the centralized international clustering solution is approximated by performing distributed clustering. To shorten the execution time, the reduction technique is then applied to remodel the proposed method into its deterministic type by replacing each unsure knowledge object with its expected centroid. Finally, the attribute-weight-entropy regularization technique enhances the proposed distributed clustering method to realize higher ends up in data clustering and extract the essential features for cluster identification. The experiments on both artificial and real-world knowledge have shown the efficiency and superiority of the presented algorithm.