-
Founta et al. Hate and Abusive Speech on Twitter
Dataset of tweets collected from 30th March 2017 to 9th April 2017 with a boosted random sampling technique, by using text analysis and preliminary crowdsourcing rounds to... -
Davidson et al. Crowd-sourced Hate Speech On Twitter Dataset
Dataset of hateful tweets sampled from Twitter using keywords. Labelled by Crowdflower, 3+ people annotated each tweet. Majority decision was taken with 92% annotator agreement. -
Fortuna et al. A Hierarchically-Labeled Portugese Hate Speech Dataset From Tw...
Dataset contains hate speech in Portuguese sampled from Twitter with 81 categories. The dataset is manually annotated for Hate Speech using a hierarchical structure of classes.... -
Ibrohim and Budi Abuse in Indonesian Twitter Dataset
Dataset of abusive tweets sampled with offensive terms. Tweets were annotated by 20 volunteer annotators and labelled by at least 3 people each. Only tweets with 100% annotators... -
Ibrohim and Budi Multi-label Hate Speech and Abusive Language Detection in In...
Dataset of hate speech and abusive language sampled from Twitter by using keywords and keyphrases. The dataset includes posts from March 2018 until September 2018 and integrated... -
Albadi et al. Arabic Religious Hate on Twitter
Dataset of Arabic religious hate tweets sampled using neutral religious names as keywords. Annotation was crowdsourced using CrowdFlower, with a minimum of 3 annotations per... -
Waseem Racism and Sexism on Twitter Dataset
Dataset of racist and sexist tweets sampled from Twitter and labelled first by experts (including feminist and anti-racist activists), and then by CF amateur annotators who... -
Ross et al. Hate Speech Against Refugees
Dataset of German annotated corpus of tweets regarding refugees in Germany. Tweets were sampled using 10 hateful hashtags and labelled by experts with 2 annotators per tweet.... -
Waseem and Hovy Racism and Sexism on Twitter Dataset
Dataset of racist and sexist tweets sampled from Twitter and labelled by a mix of expert annotators and activists. Tweets were sampled in 2016 over 2 months using keywords.... -
DKhate: Danish Hate Speech & Abusive Language data
Task description: Branching structure of tasks: Binary (Offensive, Not), Within Offensive (Target, Not), Within Target (Individual, Group, Other) Details of task:...