Social networks, in particular microblogs, have gained huge popularity in recent years. The detection of topic communities in social networks carries high value in commercial promotion, public opinion monitoring, etc. There are some existing algorithms that can detect topic communities very well. In this chapter we propose a new approach by using probabilistic generative topic model LDA (Latent Dirichlet Allocation): we add a modification to LDA to get Tri-LDA model, to process the data of friendship between users in a social network for detection of topic communities. The experiment result shows that the topic communities found by Tri-LDA are basically consistent with the realistic topic communities that are hand-labeled by the authors in the test data set.
CITATION STYLE
Ou, W., Xie, Z., Jia, X., & Xie, B. (2015). Detection of topic communities in social networks based on tri-LDA model. In Lecture Notes in Electrical Engineering (Vol. 355, pp. 1245–1253). Springer Verlag. https://doi.org/10.1007/978-3-319-11104-9_142
Mendeley helps you to discover research relevant for your work.