Incremental and adaptive topic detection over social media

Abstract

Social media like Twitter and Facebook are very popular nowadays for sharing users’ interests. However, the existing solutions on topic detection over social media overlook time and location factors, which are quite important and useful. Moreover, social media are frequently updated. Thus, the proposed detection model should handle the dynamic updates. In this paper, we introduce a topic model for topic detection that combines time and location. Our model is equipped with incremental estimation of the parameters of the topic model and adaptive window length according to the correlation of consecutive windows and their density. We have conducted extensive experiments to verify the effectiveness and efficiency of our proposed Incremental Adaptive Time Location (IncrAdapTL) model.