Implications of data density and length of collection period for population estimations using social media data

Abstract

When programmatically utilizing public APIs provided by social media services, it is possible to attain a large volume of volunteered geographic information. Geospatially enabled data from Twitter, Instagram, Panaramio, etc. can be used to create high-resolution estimations of human movements over time, with volume of the data being of critical importance. This investigation extends previous work, showing the effects of artificial data removal, and generated error; though using over twice as much collected data, attained using an enterprise cloud solution, over a span of thirteen months instead of five.