Spatio-temporal analysis of meta-data semantics of market shares over large public geosocial media data

Abstract

ABSTRACT Monitoring market share changes over space and time is an essential and continuous task for commercial companies and their third-party local agents to adjust their sale campaigns and marketing efforts for profit maximisation. This paper uses social media data as a cheap and up-to-date source to reveal the implicit semantics that are embedded in the meta-data of public geosocial datasets. We use Twitter data as a prime example of rich geosocial data. These data are associated with several meta-data attributes. Using this meta-data, we perform a geospatial analysis for the source platform from which a tweet is posted, e.g. from Apple or Android device. Our analysis studies all counties in US connected states over 2 years 2016–2017. We show that market structure at the national level masks substantial variation at the county scale. Moreover, we find strong spatial autocorrelation in platform distribution and market share in the US. In addition, we show interesting changes over the 2 years that motivates further analysis at different spatial and temporal levels. Our results are supported with visual maps of location quotients and market dominance, in addition to formal test results of spatial autocorrelation, and spatial Markov analysis.