A Weighted Network Analysis of User Migrations in a Social Network

One of the most pressing issues of our time is the negative impact of the technology towards humankind, democracy, and to nature. Impact of the social networks (and/or media) has been the center of that discussion due to well-known cases of using social networks to sway people`s opinion in events crucial to the democracy such as the election. While there is a significant public outcry about how social media is been used, as a research community, no one has looked into it objectively to understand the actual impact and the nature of the impact of social media towards elections. We collected anonymized data from Facebook surrounding the 2018 local elections. Based on that data, this paper studies the user migrations within a network of Sri Lankan political Facebook groups during local government election 2018. We find that there are users migrating among the Facebook groups during the election period. Further, we show that when the migration network, viewed as a weighted network, displays features which are significantly different from a comparable random network. In particular, we show that user migrations within Sri Lankan political groups during the election period show a non-random behavior potentially motivated by the real world events.