Identifying Online Communities of Interest using Side Information
Document Type
Conference Proceeding
Publication Date
11-6-2012
Abstract
This research investigates the potential to identify communities and individuals of interest in a weighted network by incorporating side information corresponding to the prior probability of engaging in a specific activity. A brief review of community detection techniques is presented followed by a discussion of a proposed probabilistic model for identifying communities using seeds with side information. A simulation of the model demonstrates the required parameters to detect individuals in the network who are likely to engage in a specific activity. Results highlight the ability of the model to identify small social communities by accounting for the affinity or strength of the relationships between individuals of interest and other individuals in the network.
DOI
10.1109/SSP.2012.6319641
Montclair State University Digital Commons Citation
Leberknight, Christopher; Tajer, Ali; Chiang, Mung; and Poor, H. Vincent, "Identifying Online Communities of Interest using Side Information" (2012). Department of Computer Science Faculty Scholarship and Creative Works. 327.
https://digitalcommons.montclair.edu/compusci-facpubs/327