Simultaneous community discovery and user interests extraction in social network based on probabilistic model
by Juan Bi; Zhiguang Qin; Hu Xiong; Jia Huang
International Journal of Intelligent Information and Database Systems (IJIIDS), Vol. 8, No. 3, 2014

Abstract: This article addresses the problem of discovering latent communities and topics simultaneously in social network. With the advent of online social networking, the automatic discovering communities is vital for understanding the cooperation and interaction patterns of users in these social networks. In this paper, we propose probabilistic generative models to detect latent communities by incorporating both the information of relationships and the textural content. Different from previous work, topics and user community memberships cannot be generated independently, but have a greater degree of correspondence between them. We assume that community membership is dependent on the user and a subset of topics which the user is really interested in. Furthermore, the heterogeneous relationship strengths were used to improve community discovery. These models treat community and topic as different latent variables but interdependent with each other and mutually reinforcing. Experiments on real-world dataset have shown that our models have the capability to detect well-connected and topically meaningful communities.

Online publication date: Sun, 11-Jan-2015

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Intelligent Information and Database Systems (IJIIDS):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com