Skip to content

Offensive Language Identification Dataset (OLID)

Text ClassificationHate Speech DetectionEnglish

Offensive Language Identification Dataset (OLID) is a text classification-focused dataset in English that provides 14,2 labeled examples distributed in TSV format.

About Offensive Language Identification Dataset (OLID)

Dataset contains a collection of 14,200 annotated English tweets using an annotation model that encompasses three levels: offensive language detection, categorization of offensive language, and offensive language target identification.

Details

Task
Text Classification, Hate Speech Detection
Language
English
Format
TSV
Rows / instances
14,2
Creator
Zampieri et al.
Year
2019
Download Paper

FAQ