This study addresses bias in hate speech detection by proposing a weakly supervised learning method that combines expert annotations with contrastive and prompt-based learning using large language models. The approach improves accuracy and fairness across different language groups, such as African American English and LGBTQ+ communities.
Dr. Michael Chau, Professor at HKU Business School, leads research at the intersection of AI, data analytics, and information systems, with numerous awards and editorial roles in top journals.
