HASOC (2023)

Task 1A: Identifying Hate, offensive and profane content in Sinhala

This task focuses on Hate speech and Offensive language identification for Sinhala. Sinhala is a low-resource Indo-Aryan language spoken by over 17 million people in Sri Lanka and one of the two official languages in Sri Lanka. HASOC 2023 brings the first-ever shared task for Sinhala natural language processing.

This is a coarse-grained binary classification in which participating systems are required to classify tweets into two classes, namely: Hate and Offensive (HOF) and Non- Hate and offensive (NOT).

(NOT) Non Hate-Offensive - This post does not contain any Hate speech, profane, offensive content. (HOF) Hate and Offensive - This post contains Hate, offensive, and profane content.

The train/ test sets are based on the recently released SOLD: Sinhala Offensive Language Detection dataset [1].

[1] Ranasinghe, T., Anuradha, I., Premasiri, D., Silva, K., Hettiarachchi, H., Uyangodage, L. and Zampieri, M., 2022. SOLD: Sinhala Offensive Language Dataset. arXiv preprint arXiv:2212.00851.

Task 1B: Identifying Hate, offensive and profane content in Gujarati

This task focuses on Hate speech and Offensive language identification for Gujarati. Gujarati is a low-resource Indo-Aryan language with around 50M native speakers which is one among 22 official languages in India.

This is a coarse-grained binary classification in a few shot setting, in which participating systems are required to classify tweets into two classes, namely: Hate and Offensive (HOF) and Non- Hate and offensive (NOT).

(NOT) Non Hate-Offensive - This post does not contain any Hate speech, profane, offensive content. (HOF) Hate and Offensive - This post contains Hate, offensive, and profane content.

The train set contains ~200 tweets, and participants can explore various techniques to improve the system in a few short settings.