Authors: Siming Chen, Cong Guo, Xiaoru Yuan, Jiawan Zhang, Xiaolong (Luke) Zhang
Abstract: Geo-tagged social media data can be viewed as sampling of people's trajectories in daily life. It consists of people's movements and embeds the semantics of movements. However, it is challenging to reveal patterns from the sparse and irregular sampling data. We proposed an interactive multi-filter visualization approach to analyze the spatial-temporal movement pattern in people's daily life. People's trajectories are visualized on the map with multiple functional layers. With our visual analytics tools, users are able to drill down to details, with the awareness of the origin-destination flow patterns of spatial, temporal, and semantic meaning.