Computational linguistic methods for modelling lexical-semantic dynamics of hate speech

Researcher

Sanne Hoeken

Publications

• Sanne Hoeken, Sina Zarrieß, Özge Alacam (2023): Identifying Slurs and Lexical Hate Speech via Light-Weight Dimension Projection in Embedding Space Full Publication
• Sanne Hoeken, Sophie Spliethoff, Silke Schwandt, Sina Zarrieß, Özge Alacam (2023): Towards Detecting Lexical Change of Hate Speech in Historical Data Full Publication
• Sanne Hoeken, Özge Alacam, Antske Fokkens, Pia Sommerauer (2023): Methodological Insights in Detecting Subtle Semantic Shifts with Contextualized and Static Language Models Full Publication
• Sanne Hoeken, Sina Zarrieß, Özge Alacam (2024): Hateful Word in Context Classification Full Publication
• Özge Alacam, Sanne Hoeken, Sina Zarrieß (2024): Eyes Don’t Lie: Subjective Hate Annotation and Detection with Gaze Full Publication

Collaboration

Sophie Spliethoff

Research Theme

R1 Human agency

Tags

Hate Speech LLMs

My research aims to combine work on language modelling, language change and hate speech detection to study the computational linguistic possibilities for modelling the dynamics of hate speech in social media data.