Topic Based Information Diffusion Prediction Model with External Trends

Abstract

Information diffusion model plays an important role in many real-world applications such as online marketing and e-government campaigns. Existing approaches often predict information diffusion by examining whether events are triggered by external trends or the social network itself. However, existing methods cannot take into account the semantically rich "topics" to estimate the correlations between users and messages describing some events. The main contribution of our work is the development of the Topic based Information Diffusion (TBID) model which can incorporate external trends model and topic based social descriptions to enhance the effectiveness of predicting information diffusion in online social networks. Experiments conducted based on real-world data sets confirm the distinct advantage of the proposed computational method. Our research opens the door to the development of a more effective personalized information recommendation model in online social media.