Using Social Media Content in the Visual Analysis of Movement Data

Data about the movement of people and objects is a rich source for visual analysis. However, understanding the data and inferring user behavior from it is often difficult due to missing context information. The goal of our research is to augment movement data by information derived from social media. In this paper, we present a visual concept that extends movement trajectories with terms extracted from geo-coded Twitter posts. The movement data comes from a large sample of e-bikes equipped with GPS devices. The Twitter terms are displayed as word clouds to provide additional context information for visual analysis. They are shown at locations of interests to help understand the movement data and infer possible user behavior. We plan to extend the visual concept in the future by incorporating further social media services. Ultimately, we aim for an integrated view of social media contents that allows for semi-automated reasoning and causality discovery on movement data.