What are social bots?

 

What is a social bot?

In connection with the growing use of social media, social bots have come in for criticism in recent years. Social bots are algorithmically controlled accounts that automatically produce or share content and interact with human users on social networks. By imitating human activity and behavior, social bots are hard to spot. Besides purely autonomously operating accounts, there are also hybrid accounts where human operators are partly involved and which switch back and forth between automated messages and manually created ones.
 

How do social bots work?

Social bots scan Twitter timelines for certain terms or hashtags by means of simple keyword searches. As soon as they find what they’re looking for, they comment, share links, or start a fictitious discussion. They mostly rely on predefined messages and are generally not particularly versatile in terms of their content. They can also comment directly on specific topics. Deployed in combination with other bots, their noise becomes even louder and can mislead other users.

How social bots are intended to behave is written in a suitable programming language, for example JavaScript, Python, or Ruby. The bots can then be applied on social networking platforms and are accessible through an application programming interface (API) – a kind of limited access point included by software developers in their projects which allows external parties to connect their services to them.


Categorization of social media bot accounts 

Bots can be classified into four different categories, depending on the degree of imitation of human behaviour and the distinction between benign and malicious.

     


Characteristics of social bot accounts

  • An unrealistically high number of tweets per day: Because bots work automatically, the number of posts they can produce in a single day dwarfs the output of human users. 
     
  • Profile pictures: Bots’ profile pictures often feature graphics instead of photos of real users.
     
  • Friend-follower ratio: Bot accounts generally follow far more users than they have followers, since it is far easier to follow a user than to encourage another user to follow back.
     
  • Unrealistically fast responses: Since they are based on algorithms, social bots can reply in an instant when they’re addressed.
     
  • Quality of comments: Bot accounts usually have a limited vocabulary and may produce inadequate or imprecise responses.
     
  • High number of likes: Many bot accounts award likes to users’ posts in order to receive follow-backs. This behavior results in an unusually high number of likes given compared to human accounts.
     
  • Content uniqueness: Large amounts of identical, repetitive content in posts is a sign of a bot account. Content uniqueness expresses the quantity of unique content in an account: the higher the CU, the likelier an account is to be human.

Methodology

  • The research project was headed by Prof. Stefan Stieglitz and Florian Brachten  (University of Duisburg–Essen) from 2018 to 2020.
     
  • It is one of the first studies in Germany to provide insights into bots for communication experts. 
     
  • For the first time, the researchers analyzed millions of social media posts to find out whether social bots try to influence the social media comunication of the top 30 German corporations (DAX-30 stock index). 
     
  • Secondly, in-depth interviews with representatives of companies and consultancies were conducted to find out about the scenarios in which chatbots are already used.

 

Downloads und further readings

Downloads:

Books:

  • Reeves & Nass (1996): The Media Equation: How People Treat Computers, Television, and New Media like Real People and Places.

  • Gentsch, P. (2019): AI in Marketing, Sales and Service. How Marketers without a Data Science Degree can use AI, Big Data and Bots.

Articles: 

  • Ross, B., Pilz, L., Cabrera, B., Brachten, F., Neubaum, G. & Stieglitz, S. (2019). Are social bots a real threat? An agent-based model of the spiral of silence to analyse the impact of manipulative actors in social networks. European Journal of Information System (EJIS), 28(4), pp. 394–412.
  • Bot or not? The facts about platform manipulation on Twitter. blog.twitter.com/en_us/topics/company/2020/bot-or-not.html.