Mining Human Mobility from Social Media to support Urban Computing Applications

Analysis of people trajectories is key for implementing effective urban computing applications. Nowadays, social media represent one of the main sources of information concerning human dynamics within urban context, allowing to enhance the comprehension of people behaviour, including human mobility regularities. The paper presents an approach to predict human mobility by exploiting Twitter data. The prediction method is based on a hybrid approach combining frequent pattern mining, trajectory similarity and supervised classification. The trajectory pattern similarity allows to identify the more suitable historic patterns to exploit for the prediction of the user next location. If none of the patterns satisfies the similarity threshold, a set of spatio-temporal features characterizing locations and movements among them are combined into a supervised learning approach based on M5 model trees. The experimental results obtained by using a real-world dataset show that the proposed method is effective in predicting the user's next places achieving a remarkable accuracy and prediction rate.