Hate Speech Detection Sinhala

This dataset addresses the critical issue of hate speech prevalent on social media platforms. It provides alphabetically ordered Facebook comments in Sinhala Unicode, each accompanied by a label indicating whether the comment constitutes hate speech or not. The primary purpose of this dataset is to facilitate the prediction of hateful, abusive, or insulting comments made specifically in the Sinhala language. While numerous English datasets are available for hate speech detection, this resource is particularly significant as it offers a much-needed dataset for Sinhala, the native language of Sri Lanka.

The dataset is presented as a single file, typically in a CSV format. It comprises 6,345 records. Specifically, there are 2,890 comments labelled as not hate speech (0) and 3,455 comments labelled as hate speech (1). Details regarding the specific file size in megabytes or gigabytes are not available in the provided information.

This dataset is ideally suited for developing and training machine learning models designed to detect hateful, abusive, or insulting comments on social media. Its applications include various natural language processing (NLP) tasks, classification problems, and deep learning initiatives focused on content moderation and enhancing online safety, particularly for content written in the Sinhala language.

The dataset has a global regional scope. It specifically covers comments made using Sinhala Unicode. No explicit time range or demographic details beyond the language of the comments are provided.

CC By 4.0

This dataset is intended for researchers and developers working in the fields of Artificial Intelligence and machine learning, especially those with a focus on natural language processing and deep learning for text classification. It is particularly valuable for individuals or organisations aiming to build or improve automated systems for detecting hate speech in Sinhala language content.

Original Data Source: Sinhala Unicode Hate Speech