Evaluation of graph centrality measures for tweet classification

Micro-blogging has become a very popular trend among the users of online social networks. As of January 2016, 6000 tweets are tweeted on twitter per second, Although Twitter provides a list of most popular topics people tweet about known as Trending Topics in real time, and one must have good knowledge about the current #hashtags to have the understanding of the topics. Therefore, it is interesting to explore the techniques to classify these topics into general categories with high accuracy for better information retrieval. In this paper, we explore the utilization of graph centrality measures to classify tweets accuracy. In our classifier we build graphs from the input texts, and then use properties of these graphs to classify them. We have tested the resulting technique by classifying Twitter messages in belonging to trending topic of #eq2015 and then classifying them in subtopics based on geographical information.