Multi Viewpoint Based Similarity Measure In P2p Clustering Using Pcp2p Algorithm

Abstract :

The Clustering methodologies should have some relationship among the cluster data objects. The pair of
objects in similarity will be found implicitly or explicitly. Here we introduce a novel multi viewpoint-based similarity
measure and two of its relative clustering methodologies. The Traditional dissimilarity/similarity measure is based on only a single viewpoint and that is the origin point. In our proposed methodology we introduce a different viewpoint concept even the object should not be in same cluster. Due to this we can achieve more similarity informative assessment. To prove this we are accomplishing the theoretical analysis and empirical study. We can use PCP2P algorithm for peer-to-peer clustering. The advantage of our proposed system by comparing with familiar clustering algorithms is high scalability for assigning documents to clusters.

Author Name : A.Priyadharshini, V. Kumar, R.Thiyagarajan & S. Karthikeyan

DOI: https://doi.org/10.5281/zenodo.266931

Keywords: Clustering, P2P algorithm.