vioft2nntf2t|tblJournal|Abstract_paper|0xf4ff965e270000005f22010001000100 The rise of social networks had marked the revolution and transformation of human relationships and the information age. Social networks, Facebook in specific, have more than a billion daily active users which means petabytes of data are generated every second and there are so many social interactions occurring simultaneously. Community detection revolves around the study of these social interactions and common interests to derive the most efficient method of communication to specialized groups. Considering a preferred set of features such as the posts, likes, education background and the location of users for an optimal data structure, the selection of significant users for community analysis is implemented with the unique approach to investment score and dynamic threshold allocations for the graph creation. The community detection process focuses on the analysis of cliques and map-overlay. The emphasis on the detection of overlapping communities enhances the analysis of community relationships.
Rajeswari Sridhar, Akshaya Kumar, S. Bagawathi Roshini, Ramya Kumar Sundaresan, Suganthini Chinnasamy
Anna University, Chennai, India
Community Detection, Data Structure, Link Weights, Influence Metric, Cliques, Map Overlay
| January | February | March | April | May | June | July | August | September | October | November | December |
| 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| Published By :
ICTACT
Published In :
ICTACT Journal on Soft Computing
( Volume: 8 , Issue: 4 , Pages: 1706-1713 )
Date of Publication :
July 2018
Page Views :
218
Full Text Views :
1
|