Title | ||
---|---|---|
A Novel Framework for Analyzing Overlapping Community Evolution in Dynamic Social Networks. |
Abstract | ||
---|---|---|
Finding overlapping communities from social networks is an important research topic. Previous research mainly focus on static networks, while in real world the dynamic networks are in the majority. Therefore lots of researchers turn to study dynamic social networks. One specific area of increased interest in dynamic social networks is that of identifying the critical events. However these proposed algorithms more or less exist some problems. Here in this paper we propose a novel event-based framework for analyzing overlapping community evolution in dynamic social networks. In addition, we give an index that is community tag to depict the changing process of communities over time intuitively. Moreover five indexes based on events are presented to construct the neural network prediction model, only five indexes make the complexity computation of our prediction model is simpler than the existing algorithms. Experimental results show our framework performs better, and the prediction accuracy is also acceptable. |
Year | Venue | Field |
---|---|---|
2016 | GRMSE | Data mining,Social network,Computer science,Community evolution,Artificial neural network,Computation |
DocType | Citations | PageRank |
Conference | 0 | 0.34 |
References | Authors | |
0 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Hui Jiang | 1 | 11 | 7.47 |
Xiaolong Xu | 2 | 0 | 0.34 |
Jiaying Wu | 3 | 40 | 6.00 |
Xue-Wu Zhang | 4 | 43 | 11.98 |