The objective of the task is to detect hate speech in Bengali, Bodo, and Assamese languages. It is a binary classification task. Each dataset (for the three languages) consists of a list of sentences with their corresponding class (hate or offensive (HOF) or not hate (NOT)). Data is primarily collected from Twitter, Facebook, or youtube comments. The Macro F1 score will be the yardstick of this part of the task. Team rank will be determined based on the Macro F1 score.
To get the passwords of the datasets, participants need to register first.
Registration link ----- https://forms.gle/9hS8wAeye7qiVAYo8
Download the Bengali Training dataset ---- Click here Download the Bengali Test dataset ---- Click here
Download the Assamese Training dataset ---- Click here Download the Assamese Test dataset ---- Click here
Download the Bodo Training dataset ---- Click here Download the Bodo Test dataset ---- Click here
Kaggle leaderboard links are given below
Annihilate Hates (Assamese) ---- Click here
Annihilate Hates (Bengali) ---- Click here
Annihilate Hates (Bodo) ---- Click here
Final Results are listed below
Annihilate Hates (Assamese) ---- Click here
Annihilate Hates (Bengali) ---- Click here
Annihilate Hates (Bodo) ---- Click here
18th July – training data release
25th July – test data release
15th August – run submission deadline
2nd September – results declaration
22nd September – Working notes due
15th Oct – Camera ready copies of working notes
Koyel Ghosh
Central Institute of Technology Kokrajhar, Assam, India
CVPRU, ISI Kolkata, India
Aditya Shankar Pal
CVPRU, ISI Kolkata, India
Apurbalal Senapati
Central Institute of Technology Kokrajhar, Assam, India