Thirty years of research into hate speech: topics of interest and their evolution (Scientometrics)

Social media’s explosive growth has accelerated the dissemination of misinformation and hate speech directed at specific people or groups on the basis of characteristics such as gender, color, ethnicity, or religion. International efforts have been made to define and combat hate speech as a result of this expanding problem. The current study uses co-word analysis on a 30-year-old Scopus dataset to investigate the composition and development of hate speech studies. Three primary study areas are identified by the analysis: gendered hate speech, including cyberbullying; machine learning-based detection and classification; and the dispute between hate speech and freedom of expression. Results show how important machine learning is for correctly detecting hate speech online.

https://link.springer.com/article/10.1007/s11192-020-03737-6