Introduction

This repository contains the code for the paper "Botometer 101: Social bot practicum for computational social scientists."

To demonstrate how to use Botometer for bot detection in research, we provide a simple case study and share the source code here.

How to

Obtain API keys

To replicate this case study, you will need a valid Twitter developer key with access to the V1 API. You can apply at developer.twitter.com. You will also need a valid RapidAPI account, which you can register at https://rapidapi.com. Once you have a RapidAPI account, please subscribe to a plan on https://rapidapi.com/OSoMe/api/botometer-pro/pricing . Now you are ready!

Data collection

Please refer to the scripts under folder /scripts for code and instructions to fetch data from Twitter and perform bot detection.

Data analysis

Please refer to the jupyter notebook under folder /exps for code to analyze the collected data and visualize the results.

Resources

To learn more about Botometer, you can visit the official website: https://botometer.osome.iu.edu/. Be sure to check out the FAQs.

To learn more about Botometer Pro API, you can visit https://rapidapi.com/OSoMe/api/botometer-pro/. Be sure the check out the "about" tab.

To learn more about the botometer-python package, you can visit its GitHub repo: https://github.com/IUNetSci/botometer-python.

Dependency

This demonstration uses Python3.7 and the following packages:

• botometer-python: query the Botometer Pro API
• tweepy: access Twitter API; please use versions >= 3.5.0,<4 that botometer-python depends on as well
• requests: make HTTPS requests

If you also want to replicate the analysis, you will also need

• matplotlib
• scipy

Community Guidelines

If you have questions about this project, please create issues. You are also welcome to fork the project, modify the code, and create pull requests. Please use clear and organized descriptions when creating issues and pull requests.

Cite the paper

Please cite the paper as:

@article{yang2022botometer,
  title={Botometer 101: Social bot practicum for computational social scientists},
  author={Yang, Kai-Cheng and Ferrara, Emilio and Menczer, Filippo},
  journal={Journal of Computational Social Science},
  year={2022},
  doi={10.1007/s42001-022-00177-5}
}