Optimized Clustering Process for Dynamic Data Documents

Download Now
Provided by: International Journal for Development of Computer Science & Technology (IJDCST)
Topic: Big Data
Format: PDF
A link-based algorithm which is a two stage process involving generation of a conventional matrix by discovering unknown entries through similarity between clusters in an ensemble, and then obtaining a weighted bipartite graph from this refined matrix. The authors observed the construction of the weighted bipartite graph generation is irrespective of the size of the matrix. For an optimized performance they propose to use ACO (Ant Colony Optimization) algorithm to solve minimum-weighted bipartite matching for a smaller refined matrix and metropolis algorithm for maximum-weighted bipartite matching for a larger refined matrix.
Download Now

Find By Topic