Exploring the country co-occurrence network in the twittersphere at an international economic event

Abstract

This paper explores how international relations are represented on social media in the context of an international economic event, specifically the “Belt and Road Initiative” proposed by the government of mainland China. The present study focuses on the country co-occurrence network represented in the Twittersphere, such that a link is established between two countries if they appear in the same tweet. The study also investigates how the formation of such a network can be explained by geographical, political, and economic factors. An application programming interface (API) harvested all relevant public tweets (n= 26,515) in a one-month time span (2 June–28 June 2017). The names of the countries or regions were extracted to establish the network, with 52 nodes (countries or regions) and 86 edges. Social network analysis revealed that mainland China, Hong Kong, Pakistan, Greece, Kenya, and Iran were in the network’s important positions, as indicated by their high betweenness centrality. Exponential random graph modeling (ERGM) results suggested that West Asian countries engaging heavily in international polities, countries with lower levels of press freedom, and those receiving less direct investment from mainland China, were more likely to be tweeted together.