Mobility in cities: Comparative analysis of mobility models using Geo-tagged tweets in Australia

Modeling human movement has recently received significant research interest due to its applications in disaster management, transportation planning, communication networks and epidemic modeling and prediction. Most of the prior human mobility studies use data which is coarse-grained, proprietary or limited in size. In contrast, this study utilizes a large amount of publicly available data from Twitter. Geo-data associated with Twitter posts have been analyzed to characterize and model movement patterns within Australia. Gravity and Radiation models have been used to analyze human movements within major Australian cities. The Gravity models show better performance compared to the Radiation model in estimating flows between places. We further find that fitted parameters vary across cities, highlighting the need for city-specific models to accurately represent movement flows.