A large-scaled dataset for Vietnamese Hate Speech Detection on Social media texts. The dataset is crawled from Facebook and Youtube, and is manually annotated by human.

The dataset addresses discriminations across sexuality, ethnicity, and gender. Posts are sampled slur-based ("f*ggot","n*gger","tr*nny") from several subreddits from October...

Dataset of hateful Wikipedia comments. The sampling of the data was a combination of random + oversampled on banned comments. Annotation was crowdsourced, and each comment was...

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...

The repo contains a list of words that Shutterstpck uses to filter results from our autocomplete server and recommendation engine. Can be installed in a npm project by: npm...

YouTube Blacklist Words List includes; a list of unacceptable words, inappropriate words, a list of swear words, offensive words, curse words, insulting words, all cuss words,...

WordPress Comment Blacklist Words, WordPress Comment Moderation, and WordPress Comment Spam The WordPress Comment Blacklist Words/Phrases include; a list of swears, unacceptable...

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.

Datasets contain racism and sexism in Sinhalese from Twitter. The data was sampled using pre-identified keywords from surveys and experts. The data was annotated by experts...

This dataset expands the OLID/OffensEval (OLID (Zampieri et al., 2019a), Offensive Language Identification Dataset) by adding the explicitness of the message. The OLID data was...

Dataset of self-reported microaggressions from microaggressions.com. 2,934 posts were collected targeted towards gender (1,314 posts), race (1,278 posts), sexuality (461 posts),...

Dataset of hate speech in Indonesian, including hatred for religion, race, ethnicity, and gender. Posts from Twitter are sampled using relevant hashtags to contentious political...

Dataset of Hinglish sexist Tweets sampled by crawling popular hashtags and well-known people. Tweets were labelled by experts, with an average Cohen's kappa of 0.83.

Dataset of moderated news comments from Slovene RTV MCC. Comments were labelled by expert annotators based on the type of inappropriate content. Note that this data is encrypted.

Dataset of moderated news comments from Croatian 24sata. Comments were labelled by expert annotators based on the type of inappropriate content. Note that this data is encrypted.

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....

Dataset 1 includes offensive Arabic tweets sampled in March 2014 using obscene keywords and hashtags used for pornographic pages (available as a .txt file word list). Dataset 2...

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...

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...

Datasets contain offensive comments from YouTube. The data was sampled from 2015 to 2017 and collected in July 2017. Channels with controversial comments about celebrities were...

Dataset collected from the World of Warcraft (dataset 1) and League of Legends (dataset 2) forum. 20 topics were selected for each dataset based on offensive terms from...

Dataset of Hate Speech and Abusive language. It is streamed from Twitter from March 2018 to February 2019 using relevant terms and from specified users, which have 100k...

Dataset of Facebook posts and comments published in response to them from the Facebook pages “Pegida” (dataset 1), “Ich bin Patriot, aber kein Nazi” (“I’m a patriot, not a...

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...

Dataset of sexist tweets sampling based on benevolent sexist key phrases from which 712 tweets were manually selected by the authors, and validated by three non-activist...