Multi-resolution Spatial Event Forecasting in Social Media

Liang Zhao, Feng Chen, Chang-Tien Lu, Naren Ramakrishnan

Abstract

Social media has become a significant surrogate for spatial event forecasting. The accuracy and discernibility of aspatial event forecasting model are two key concerns, which respectively determine how accurate and how detailed the model's predictions could be. Existing work pays most attention on the accuracy alone, seldom considering the accuracy and discernibility simultaneously, because this would requires a considerably more sophisticated model while still suffering from several challenges: 1) the precise formulation of thetrade-off between accuracy and discernibility, 2) the scarcity of social media data with a high spatial resolution, and 3)the characterization of spatial correlation and heterogeneity. This paper proposes a novel feature learning model that concurrently addresses all the above challenges by formulating prediction tasks for different locations with different spatialresolutions, allowing the heterogeneous relationships among the tasks to be characterized. This characterization is then integrated into our new model based on multitask learning, whose parameters are optimized by our proposed algorithm based on the Alternative Direction Method of Multipliers(ADMM). Extensive experimental evaluations on 11 data sets from different domains demonstrated the effectiveness of our proposed approach.