HateThaiSent Dataset and ThaiCaps multi-tasking framework

Krishanu Maity, A.S. Poornash, Shaubhik Bhattacharya, Salisa Phosit, Sawarod Kongsamlit, Sriparna Saha, and Kitsuchart Pasupa

---

This repository contains annotated data of Thai hate speech available for hate speech detection and sentiment analysis and also the proposed multi-task deep learning framework ThaiCaps along with its SOTA counterparts for comparison, which appeared in the paper "HateThaiSent: Sentiment-Aided Hate Speech Detection in Thai Language"

Details

• The dataset contains 7597 messages.
• The annotations are contained in the format of a CSV file.
• Each message was annotated by the hateful Hatespeech, Nonhatespeech and the sentiment Positive, Neutral, Negative of the message.
• The Baselines folder contains the compared SOTA approaches and the ThaiCaps folder contains the proposed framework.
• Each framework has a run.py file which can be executed with -

  CUDA_VISIBLE_DEVICES=0 python run.py

Licensing

This dataset is licensed under the MIT License.

Citation

• Krishanu Maity, A.S. Poornash, Shaubhik Bhattacharya, Salisa Phosit, Sawarod Kongsamlit, Sriparna Saha, and Kitsuchart Pasupa (2024). HateThaiSent: Sentiment-Aided Hate Speech Detection in Thai Language.

@article{HateThaiSent,
    author = {Krishanu Maity, A.S. Poornash, Shaubhik Bhattacharya, Salisa Phosit, Sawarod Kongsamlit, Sriparna Saha, and Kitsuchart Pasupa},
    title = {{HateThaiSent: Sentiment-Aided Hate Speech Detection in Thai Language}},
    journal = {IEEE Transactions on Computational Social Systems},
    year = {2024},
}