Multi-Topic Tracking Model for dynamic social network

Yuhua Li, Changzheng Liu, Ming Zhao, Ruixuan Li, Hailing Xiao, Kai Wang, Jun Zhang

Research output: Contribution to journalArticlepeer-review

5 Scopus citations


The topic tracking problem has attracted much attention in the last decades. However, existing approaches rarely consider network structures and textual topics together. In this paper, we propose a novel statistical model based on dynamic bayesian network, namely Multi-Topic Tracking Model for Dynamic Social Network (MTTD). It takes influence phenomenon, selection phenomenon, document generative process and the evolution of textual topics into account. Specifically, in our MTTD model, Gibbs Random Field is defined to model the influence of historical status of users in the network and the interdependency between them in order to consider the influence phenomenon. To address the selection phenomenon, a stochastic block model is used to model the link generation process based on the users' interests to topics. Probabilistic Latent Semantic Analysis (PLSA) is used to describe the document generative process according to the users' interests. Finally, the dependence on the historical topic status is also considered to ensure the continuity of the topic itself in topic evolution model. Expectation Maximization (EM) algorithm is utilized to estimate parameters in the proposed MTTD model. Empirical experiments on real datasets show that the MTTD model performs better than Popular Event Tracking (PET) and Dynamic Topic Model (DTM) in generalization performance, topic interpretability performance, topic content evolution and topic popularity evolution performance.

Original languageEnglish (US)
Pages (from-to)51-65
Number of pages15
JournalPhysica A: Statistical Mechanics and its Applications
StatePublished - Jul 15 2016


  • Dynamic social network
  • Influence phenomenon
  • Multi-Topic Tracking Model
  • Selection phenomenon

ASJC Scopus subject areas

  • Statistical and Nonlinear Physics
  • Statistics and Probability


Dive into the research topics of 'Multi-Topic Tracking Model for dynamic social network'. Together they form a unique fingerprint.

Cite this