Can Twitter be a Reliable Proxy to Characterize Nation-wide Human Mobility? A Case Study of Spain

Abstract

In recent years, there has been an increasing interest in the exploitation of geo-tagged documents posted on Online Social Networks (OSN) for human-mobility pattern mining. These patterns can ease the development of effective and intelligent location-based systems of different scenarios. However, the validation of OSN geo-data as a reliable source for human mobility has not been fully studied in literature. Therefore, this study proposes a comprehensive comparison of a nation-scale Twitter (TWT) dataset with an official mobility study published by the Spanish Ministry of Development. Both feeds have been compared considering different variables, such as population density and spatial granularity, among others. Results show that TWT can be a reliable source for human-mobility mining but only when certain socioeconomic, temporal, and spatial factors co-occur.