Identifying and tracking topic-level influencers in the microblog streams (Q1640397)

This paper deals with the topic-level temporal influence of users for determining those popular and influential persons on specific topics in the microblog sphere. The paper proposes a topic-level influence over time (TIT) model, which is a probabilistic generative model, where the users co-followed by others with similar topical interests can be a good indicator of key influencers on the topics. The model groups the topic independent links into another cluster as the topic unrelated influencers because the topic is not the only reason for users to follow other in microblog and Bernoulli distribution is used to capture the attention to another on microblog. The influence decay is also taken into consideration in the computation of the topic-level influence of each user. Extensive simulations have been conducted based on real-world data sets to illustrate the effectiveness of the proposed methods.