Virtual Booth

MYHATE is a multilingual hate speech detection framework designed to address the growing spread of harmful online content within the Malaysian social media landscape. The innovation focuses on detecting hate speech related to the highly sensitive 3R elements (Race, Religion, and Royalty) which are often overlooked by existing systems that rely heavily on English-centric datasets. MYHATE integrates a complete detection pipeline, including social media data collection, dual-layer annotation, text preprocessing, and model development using both machine learning and deep learning approaches. By incorporating Malaysia-specific linguistic characteristics in both Malay and English, the framework enhances contextual understanding and detection accuracy. The system’s performance is evaluated using standard metrics such as accuracy, precision, recall, and F1-score, demonstrating its effectiveness on unseen data. Overall, MYHATE offers a culturally adaptive, scalable, and practical solution for monitoring and mitigating hate speech, supporting safer digital discourse and responsible social media governance in Malaysia