Predicting Popularity of Forum Threads for Public Events Security

Web user's online interactive behavior with others often makes some user generated contents popular. The modeling and prediction of the popularity of online content are an important research issue for many key application domains. In this paper, we focus on one form of user generated content, forum threads, and their popularity prediction for public events security. To predict the popularity of forum threads, we first define the popularity prediction problem, and identify the dynamic factors that affect the popularity of forum threads. Based on the information of dynamic evolution at the early stage, we propose a popularity prediction algorithm which makes use of the locality property and combines multiple dynamic factors. The proposed algorithm is further evaluated using the Tianya forum dataset on the discussions of various public events. The experimental results show that, compared to the baseline methods, our method achieves relatively better performance in predicting the popularity of forum threads on public events security.