A Survey on Detecting Hate Speech and Misogyny in Native and Code-Mixed Texts in Social Media
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
The wide usage of social media has transformed global communication, but also amplified the dissemination of hate speech and misogynistic content that is prone to severe threats to online safety and societal harmony.The usage of code-mixed texts, where users mix English with regional languages, has increased in multilingual countries like India.The complex semantics, phonological variances, and sociolinguistic indicators included in such multilingual expressions are mostly failed by traditional hate speech detection methods.This review article presents a complete investigation of existing approaches for detecting hate speech and misogyny in native and codemixed languages.The study explores various embedding approaches, such as static and contextual transformer-based multilingual embeddings and recurrent neural networks.Attention mechanisms are used for better semantic understanding in low-resource languages, which is also discussed.This study promotes the importance of developing robust, inclusive, and linguistically adaptive models for effective detection of hate and misogyny in multilingual online ecosystems.
Fields of Study & Subjects
Hate Speech and Cyberbullying Detection
Sentiment Analysis and Opinion Mining
Misinformation and Its Impacts
Publication details
- Journal / source
- Advances in intelligent systems research/Advances in Intelligent Systems Research
- Pages
- 641–650
- Publication date
- 1 April 2026
- Language
- EN
- Document type
- Book chapter