Learning Social Meta-knowledge for Nowcasting Human Mobility in Disaster

What is it about?

Nowcasting human mobility in disaster is the most topic for disaster management and response. However, human mobility under big disasters such as hurricanes and pandemics deviates from its daily routine to a large extent, which makes the task more challenging. To solve this problem, we utilize tweet count as an effective covariate for human mobility nowcasting during disasters and propose a novel multimodal learning framework called Meta-knowledge Memorizable Spatio-Temporal Network (MemeSTN) to learn the cross-modal correlations and inter-dependencies between the social media and mobility data.

Featured Image