Hugging Face Readme: BanglaHateBERT

License: OpenRail

This repository contains the BanglaHateBERT model, designed for Abusive Language Detection in Bengali. It leverages the powerful BERT architecture and is made available through the Hugging Face Transformers library.

Installation

To use BanglaHateBERT, you need to install the transformers library:

pip install transformers

Usage

Here's a simple example demonstrating how to load the tokenizer and the model:

from transformers import AutoTokenizer, AutoModelForSequenceClassification

tokenizer = AutoTokenizer.from_pretrained("saroarj/BanglaHateBert")
model = AutoModelForSequenceClassification.from_pretrained("saroarj/BanglaHateBert")

Citation

If you use BanglaHateBERT in your research or work, please cite the following paper:

@inproceedings{jahan2022banglahatebert,
  title={BanglaHateBERT: BERT for Abusive Language Detection in Bengali},
  author={Jahan, Md Saroar and Haque, Mainul and Arhab, Nabil and Oussalah, Mourad},
  booktitle={Proceedings of the Second International Workshop on Resources and Techniques for User Information in Abusive Language Analysis},
  pages={8--15},
  year={2022}
}

Test Dataset and Source Code

You can find the test dataset for BanglaHateBERT at the following link: https://github.com/saroarjahan/BanglaHateBert.

For the source code, please email me at [saroarjahan01[a..t]gmail . com].