Geo-located Twitter as proxy for global mobility patterns.

Area of Science:

• Computational social science
• Mobility studies
• Network analysis

Background:

• Location-sharing services provide unprecedented data on human activity.
• Understanding global mobility patterns is crucial for various fields.

Purpose of the Study:

• To analyze geo-located Twitter messages to uncover global human mobility patterns.
• To estimate international traveler volumes and examine national mobility profiles.
• To identify temporal patterns and regional structures in global mobility.

Main Methods:

• Analysis of nearly one billion geo-located tweets from 2012.
• Estimation of international traveler volume by country of residence.
• Examination of mobility rate, radius of gyration, destination diversity, and inflow-outflow balance.
• Community structure analysis of the Twitter mobility network.
• Validation using global tourism statistics and existing mobility models.

Main Results:

• Estimated international traveler volumes by country of residence.
• Identified distinct mobility profiles for different nations.
• Revealed universal seasonal patterns of international mobility.
• Discovered spatially cohesive regions mirroring global regional divisions.
• Validated Twitter data as a reliable source for mobility studies.

Conclusions:

• Geo-located tweets offer a powerful tool for understanding and quantifying global human mobility.
• Twitter data reveals significant insights into international travel patterns and regional mobility structures.
• Findings are consistent with global tourism statistics and mobility models, underscoring the utility of social media data.