Abusive Language Detection on Arabic Social Media

By Hamdy Mubarak, Kareem Darwish, Walid Magdy. 2017

320 citations recorded by the source corpus.

In this paper, we present our work on detecting abusive language on Arabic social media. We extract a list of obscene words and hashtags using common patterns used in offensive and rude communications. We also classify Twitter users according to whether they use any of these words or not in their tweets. We expand the list of obscene words using this classification, and we report results on a newly created dataset of classified Arabic tweets (obscene, offensive, and clean). We make this dataset freely available for research, in addition to the list of obscene words and hashtags. We are also publicly releasing a large corpus of classified user comments that were deleted from a popular Arabic news site due to violations the site’s rules and guidelines.

Subfields: Artificial Intelligence, Communication, Law, Natural Language Processing, Safety, Robustness & Interpretability

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Communication

In Law

In Natural Language Processing

In Safety, Robustness & Interpretability

Citations in the corpus

Language Generation Models Can Cause Harm: So What Can We Do About It? An Actionable Survey (2022, background): “Since language is inherently a tool of power—the primary means by *Equal contribution 1While the majority of these models are trained on English language datasets, recent studies have also obtained similar gains on other languages (Lin et al., 2021c; Shliazhko et al., 2022) which people and societie”
Predicting the Type and Target of Offensive Posts in Social Media (2019, background): “Most prior work focuses on a different aspect of offensive language such as abusive language (Nobata et al., 2016; Mubarak et al., 2017), (cyber-)aggression (Kumar et al., 2018), (cyber-)bullying (Xu et al., 2012; Dadvar et al., 2013), toxic comments , hate speech (Kwok and Wang, 2013; Djuric et…”
Resources and benchmark corpora for hate speech detection: a systematic review (2020, background): “Mubarak et al. (2017) (MDM)—three resources for Arabic language including: a lexicon of 288 obscene words; a test set of 1100 tweets for manual validation; a dataset of 32,000 comments that have been removed from the popular news website AlJazeera.”
Directions in Abusive Language Training Data: Garbage In, Garbage Out (2020, background): “In some cases, specific types of group-directed abuse are subsumed within a broader category of identity-directed abuse, as in [4, 78, 119].”
Introducing CAD: the Contextual Abuse Dataset (2021, background): “A well-established problem with abusive content datasets is that each bit of content is marked up individually, without taking into account any content that came before (Gao and Huang, 2017; Mubarak et al., 2017).”
Overview of the HASOC track at FIRE 2019: Hate Speech and Offensive Content Identification in Indo-European Languages (2019, background): “Arabic Social Media [22] Twitter, Arabic 1100 tweets, 32000 comments Obscene, inappropriate F1 around 0.”
Tackling Online Abuse: A Survey of Automated Abuse Detection Methods (2019, methodology): “Working with the DATA-GAZZETTA dataset, Pavlopoulos et al. (2017c) incorporated user embeddings into Pavlopoulos’ setup 1 (2017a; 2017b) described above.”
Detecting Hate Speech in Social Media (2017, background): “Mubarak et al. (2017) addresses abusive language detection on Arabic social media and Su et al. (2017) presents a system to detect and rephrase profanity in Chinese.”
Toxicity Detection can be Sensitive to the Conversational Context (2021, background): “, 2017a), Arabic (Mubarak et al., 2017), French (Chiril et al.”
Toxic Language Detection in Social Media for Brazilian Portuguese: New Dataset and Multilingual Analysis (2020, background): “Arabic Mubarak et al. (2017) present a dataset with 1 .”
XHate-999: Analyzing and Detecting Abusive Language Across Domains and Languages (2020, unknown): “A non-exhaustive list includes Arabic (Mubarak et al., 2017; Chowdhury et al., 2020), Danish (Sigurbergsson and Derczynski, 2020), German (Jaki and De Smedt, 2019; Wiegand et al., 2018b), Hindi (Saroj and Pal, 2020), Italian (Bosco et al., 2018; Fersini et al., 2018), Polish (Ptaszynski et al.,…”
Context Sensitivity Estimation in Toxicity Detection (2021, background): “Most of the datasets in the ﬁeld are in English, but datasets in other languages have the same weakness (Pavlopoulos et al., 2017a; Mubarak et al., 2017; Chiril et al., 2020; Ibrohim and Budi, 2018; Ross et al., 2016; Wiegand et al., 2018).”