A Novel Algorithm for Top- k Community Detection In Dynamic Social Networks

Abstract

In dynamic informal networks change the information prevailing inside them in perspective with time. An approach to distinguish the information existed in each nodes was achieved by recognizing the subnetworks or Communities inside the network through Community-Based models. To increase the utility capacity also estimated the degree to networks that are founded. The information about the networks will be not known and consequently the nature of the topology and instability data inside a Dynamic Social Network(DSN). It maximized the cost of finding the communities from the large networks and to reduce the cost expenditure in finding the communities; a novel greedy based approach has been proposed. The K values determine the total number of communities that are to be formed. The Network-based model foresee the individuals from the networks proposed strategy will likewise distinguish the network -based model that can catch the qualities of basic opening spanners productively and decline the connection upkeep cost in the extensive scale systems. An decrease in the cost of finding the optimal communities in comparison with the existing hierarchical and Clique census community detection approaches was achieved by the Top-k greedy approach with an optimal cost percent of 42.6.