Social Bots

An example visualization showing a concerning pattern of heavy bot follower activity for a self-identified journalist, particularly for the first 30.000 followers [2].

I am working on this project with Onur Varol, studying social bots, their detection, and implications of their existence. We have studied deception strategies for online discussions drawing parallels from history[1], and the relationship between social bots and journalists on twitter[2]. We are currently conducting a validation study on the publicly available social bot detection tools.

[1] - Varol, O., & Uluturk, I. (2018). Deception strategies and threats for online discussions. First Monday, 22(5). Download
[2] - Varol, O., & Uluturk, I. (2019). Journalists on Twitter: Self-branding, Audiences, and Involvement of Bots. Journal of Computational Social Science. Link

We are also developing social-annotate, an open-sourced tool to simplify data annotation and collection from social media sites. We are taking great care to avoid isolating the annotation task from the regular use of the site, which allows users to collect and annotate data while experiencing the site normally. This also makes studying the user experience on the platform itself possible. One our main goals is to make this tool usable for researchers lacking technical know-how.