This repository contains a trained model for detecting Twitter bot posts, akin to the functionality of Botometer. Given that Botometer is a premium service requiring payment for API access, this project has been developed as an open-source alternative.
Our model has been trained on 50,000 bot and genuine posts using state-of-the-art machine learning technologies, such as OpenAI and BERT. It has achieved an overall accuracy of 87% in bot detection.
The dataset used for training was sourced from open-access resources, and benchmarked data was employed for testing purposes. One such dataset used was TwiBot-20.
Reference papaer:
@inproceedings{feng2021twibot, title={Twibot-20: A comprehensive twitter bot detection benchmark}, author={Feng, Shangbin and Wan, Herun and Wang, Ningnan and Li, Jundong and Luo, Minnan}, booktitle={Proceedings of the 30th ACM International Conference on Information & Knowledge Management}, pages={4485--4494}, year={2021} }
The trained model, model_1.pt, is available for download from Google Drive via the following link. Should you encounter any issues in accessing the model, please feel free to contact me via email at saroarjahan.bd@gmail.com.
To use the model, run the model.py or model.ipynb file, and make sure u have downloawd model_1.pt bfeore running it.
You may need to install the following libraries:
torch: Install usingpip install torchtransformers: Install usingpip install transformerspandas: Install usingpip install pandas
These libraries are necessary for running the model and processing data.