Abstract
Abstract. It is difficult to quantify how – and to what extent – the public engages with events in other countries. Twitter users all over the globe post more than 500 million tweets every day. They also discuss places in their tweets. Therefore, Twitter provides a lens through which geographic research can investigate public discourse as it relates to place. Further, many research studies use geo-tagged posts on Twitter (and social media in general) to sense the society in particular locations (according to geo-tags) for various purposes that may need a “local sense” such as sentiment analysis, situational awareness for crisis response, election prediction, or targeted advertising. However, it is unclear to what extent the online discourse by users are about local events versus events in other locations or countries.In this pilot study, we visualize and characterize relations between places mentioned in Twitter posts and places where users live to identify whether Twitter users in different countries engage more with domestic or international (or transnational) events. We also visualize the extent to which places in other countries are being discussed through online platforms/social media. The results have implications for the design of algorithms in geographic information science attempting to automatically geolocate places mentioned in tweets for use in sentiment or spatial analysis, situational awareness, and advertisement. Additionally, most place names are ambiguous and refer to more than one location. For example, London can refer to London, Texas or London in England. Our analysis gauge whether Twitter users’ profile locations can be used to disambiguate places that are mentioned in their tweets.
Highlights
In this pilot study, we visualize and characterize relations between places mentioned in Twitter posts and places where users live to identify whether Twitter users in different countries engage more with domestic or international events
We used the GeoCorpora dataset (Wallgrün, 2018) to conduct our study using a gold-standard dataset, i.e. a manually geo-annotated dataset in which place names were manually resolved to geolocations, since according to our analysis, the accuracy of automated text geoparsing methods usually does not exceed 80% for tweets
We processed the resulting 2185 tweets that had place names in their textual content to establish a link between every place listed in the profile location and every place mentioned in the tweet text
Summary
We visualize and characterize relations between places mentioned in Twitter posts and places where users live to identify whether Twitter users in different countries engage more with domestic or international (or transnational) events. To generate this corpus, crisisrelated keywords were used to filter for tweets that were more likely to include place names.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
