[Paper Review] Do Social Bots Dream of Electric Sheep? A Categorisation of Social Media Bot Accounts

Stefan Stieglitz, Florian Brachten+2 more|arXiv (Cornell University)|Oct 11, 2017

Spam and Phishing Detection34 references108 citations

TL;DR

The paper conducts a systematic literature review to define social bots on social media and proposes a two-dimensional categorisation scheme based on Imitation of human behaviour and Intent.

ABSTRACT

So-called 'social bots' have garnered a lot of attention lately. Previous research showed that they attempted to influence political events such as the Brexit referendum and the US presidential elections. It remains, however, somewhat unclear what exactly can be understood by the term 'social bot'. This paper addresses the need to better understand the intentions of bots on social media and to develop a shared understanding of how 'social' bots differ from other types of bots. We thus describe a systematic review of publications that researched bot accounts on social media. Based on the results of this literature review, we propose a scheme for categorising bot accounts on social media sites. Our scheme groups bot accounts by two dimensions - Imitation of human behaviour and Intent.

Motivation & Objective

• Clarify what constitutes a social bot on social media platforms.
• Summarize existing research to distinguish social bots from other bot types.
• Develop a shared, practical categorisation framework for bot accounts.
• Highlight how bot behavior and intent relate to manipulation of social media dynamics.

Proposed method

• Perform a systematic literature review of publications researching bot accounts on social media.
• Extract and synthesize common dimensions used to describe social bots.
• Propose a categorisation scheme organized along two axes: Imitation of human behaviour and Intent.

Experimental results

Research questions

• RQ1 What characteristics define social bots on social media compared to other bots?
• RQ2 How can social bots be differentiated based on behavior imitation and underlying intent?
• RQ3 What a priori scheme can capture variations across bot accounts found in literature?

Key findings

• Bot accounts on social media have varied levels of human-like behaviour and explicit vs. implicit intent.
• A two-dimensional framework effectively distinguishes bot accounts by Imitation (degree of human-like actions) and Intent (purpose behind actions).
• Existing studies show bots can influence online discourse, reinforcing the need for a shared understanding of categorisation.