Spatial-temporal analysis of international connections based on textual social media data

Abstract

The development of social media has improved public discussion online. The proliferation of social media data provides easy access to a huge amount of user-generated content. Geotagged microblogs have been investigated in a variety of research fields for understanding spatial temporal pattern in subjects like human mobility and characteristics of a city. Besides geotagged microblogs, textual content can also be used to harvest geographical information. However, there is little research on this topic. Therefore, in this study, we aim to demonstrate how we can illustrate the connections between a virtual community and other states and regions by analysing microblogs. To fulfill the research aim, we present the workflow of how to define a virtual community and harvest geographical information from microblog feeds. Then we discuss the visualization techniques we need to display such information. In order to do so, Sina Weibo was picked as a data source to demonstrate the international connections between China and rest of the world. Through visual analysis, we find out that even in Cyberspace, the physical world still has an impact, and it can be proved that states geographically closer to China have a higher level of attention (stronger data flow). In the meantime, we can also notice the barrier of information exchange caused by real space distance has less influence in Cyberspace, a virtual community can have strong bound with faraway country. But generally speaking, the level of attention of a state is related to its economy power. And there is a different topic preference between male and female user. This study may provide a new research aspect for analysis human thought, the spread of news, international relations, cross-culture communication based on textual geoinformation contained in social media data.