On power-law relationships of the Internet topology
Proceedings of the conference on Applications, technologies, architectures, and protocols for computer communication
The author-topic model for authors and documents
UAI '04 Proceedings of the 20th conference on Uncertainty in artificial intelligence
Introduction to Data Mining, (First Edition)
Introduction to Data Mining, (First Edition)
Identifying the influential bloggers in a community
WSDM '08 Proceedings of the 2008 International Conference on Web Search and Data Mining
Predicting tie strength with social media
Proceedings of the SIGCHI Conference on Human Factors in Computing Systems
A measurement-driven analysis of information propagation in the flickr social network
Proceedings of the 18th international conference on World wide web
Mining topic-level influence in heterogeneous networks
CIKM '10 Proceedings of the 19th ACM international conference on Information and knowledge management
Probabilistic latent semantic analysis
UAI'99 Proceedings of the Fifteenth conference on Uncertainty in artificial intelligence
Hi-index | 0.00 |
Large social networks (e.g. Twitter, Digg and LinkedIn), have successfully facilitated information diffusion related to various topics. Typically, each topic discussed in these networks is associated with a group of members who have generated content on it and these users form a topic community. Tracking topic community is of much importance to predict the trend of hot spots and public opinion. In this paper, we formally define the problem of topic community tracking as a two-step task, including topic interest modeling and topic evolution mining. We proposed a topic community tracking model to model user's interest on a topic which is based on random walk algorithm and combines user's personal affinity and social influence. And then, considering that user's interest on a topic will vary with time when the discussion content changes and new topic community member joins in, we explore an Adaptive Topic Community Tracking Model. Comprehensive experimental studies on Digg and Sina Weibo corpus show that our approach outperforms existing ones and well matches the practice.