Automated Hate Speech Detection and the Problem of Offensive Language
arXiv:1703.04009
Abstract
A key challenge for automatic hate-speech detection on social media is the separation of hate speech from other instances of offensive language. Lexical detection methods tend to have low precision because they classify all messages containing particular terms as hate speech and previous work using supervised learning has failed to distinguish between the two categories. We used a crowd-sourced hate speech lexicon to collect tweets containing hate speech keywords. We use crowd-sourcing to label a sample of these tweets into three categories: those containing hate speech, only offensive language, and those with neither. We train a multi-class classifier to distinguish between these different categories. Close analysis of the predictions and the errors shows when we can reliably separate hate speech from other offensive language and when this differentiation is more difficult. We find that racist and homophobic tweets are more likely to be classified as hate speech but that sexist tweets are generally classified as offensive. Tweets without explicit hate keywords are also more difficult to classify.
To appear in the Proceedings of ICWSM 2017. Please cite that version
Cited by in corpus (40)
- Stereotypical Bias Removal for Hate Speech Detection Task using Knowledge-based Generalizations
- The Structure of Toxic Conversations on Twitter
- An Online Multilingual Hate speech Recognition System
- Explainable AI: current status and future directions
- Detecting Hate Speech in Multi-modal Memes
- "Like Sheep Among Wolves": Characterizing Hateful Users on Twitter
- Hateminers : Detecting Hate speech against Women
- Detecting Hate Speech in Social Media
- Trajectories of Blocked Community Members: Redemption, Recidivism and Departure
- Interpretable Multi-Modal Hate Speech Detection
- Misogynistic Tweet Detection: Modelling CNN with Small Datasets
- Offensive Language and Hate Speech Detection with Deep Learning and Transfer Learning
- Towards generalisable hate speech detection: a review on obstacles and solutions
- A large-scale crowdsourced analysis of abuse against women journalists and politicians on Twitter
- Hostility Detection Dataset in Hindi
- Addressing machine learning concept drift reveals declining vaccine sentiment during the COVID-19 pandemic
- SWE2: SubWord Enriched and Significant Word Emphasized Framework for Hate Speech Detection
- Understanding and Detecting Dangerous Speech in Social Media
- Designing Toxic Content Classification for a Diversity of Perspectives
- Walk in Wild: An Ensemble Approach for Hostility Detection in Hindi Posts
- Lifelong Learning of Hate Speech Classification on Social Media
- HASOCOne@FIRE-HASOC2020: Using BERT and Multilingual BERT models for Hate Speech Detection
- Learning from Fact-checkers: Analysis and Generation of Fact-checking Language
- How Hateful are Movies? A Study and Prediction on Movie Subtitles
- Low-Shot Classification: A Comparison of Classical and Deep Transfer Machine Learning Approaches
- Reading Between the Demographic Lines: Resolving Sources of Bias in Toxicity Classifiers
- Detecting Inappropriate Messages on Sensitive Topics that Could Harm a Company's Reputation
- Tackling Racial Bias in Automated Online Hate Detection: Towards Fair and Accurate Classification of Hateful Online Users Using Geometric Deep Learning
- Detecting Hostile Posts using Relational Graph Convolutional Network
- A survey on extremism analysis using Natural Language Processing
- Effect of Word Embedding Models on Hate and Offensive Speech Detection
- Analysis of the Ethiopic Twitter Dataset for Abusive Speech in Amharic
- IIITG-ADBU@HASOC-Dravidian-CodeMix-FIRE2020: Offensive Content Detection in Code-Mixed Dravidian Text
- Characterizing Abhorrent, Misinformative, and Mistargeted Content on YouTube
- A Bootstrapped Model to Detect Abuse and Intent in White Supremacist Corpora
- Statistical Analysis of Perspective Scores on Hate Speech Detection
- Sexism Identification in Tweets and Gabs using Deep Neural Networks
- Anomaly-Injected Deep Support Vector Data Description for Text Outlier Detection
- Reducing Unintended Bias of ML Models on Tabular and Textual Data
- Distributionally Robust Removal of Malicious Nodes from Networks