Predicting the topic influence trends in social media with multiple models

Online social networks, such as twitter and facebook, are continuously generating the new contents and relationships. To fully understand the spread of topics, there are some essential but remaining open questions. Why are some seemingly ordinary topics attracting? Is it due to the attractiveness of the content itself, or some external factors, such as network structure, time or event location, play a larger role in the dissemination of information? Analyzing the influence and spread of upcoming contents is an interesting and useful research direction, and has brilliant perspective on web advertising and spam detection. In this paper, a novel time series model for predicting the topics social influence has been introduced. In this model, the existing user-generated contents are summarized with a set of valued sequences, and a hybrid model consisting of topical, social and geographic attributes has been adopted for predicting influence trends of newly coming contents. The empirical study conducted on large real data sets indicates that our model is interesting and meaningful, and our methods are effective and efficient in practice.