معرفی
Dr. Hansi Hettiarachchi is a Lecturer in Security and Protection Science at the Department of Computing and Communications, Lancaster University. Their research focuses on developing Machine Learning approaches for Natural Language Processing (NLP) tasks, particularly emphasizing societal and human security and safety. Key areas include online event detection (e.g., temporal profiling, trigger/argument detection), online safety (offensive content detection, misinformation identification), and information extraction (named entity recognition, rule generation). They also explore Large Language Models' capabilities, multilingual NLP, low-resource language support, and model explainability.
Recent research includes creating datasets like SOLD (Sinhala Offensive Language Dataset) and CODE-ACCORD (building regulatory corpus), organizing the LoResLM workshop on low-resource languages, and advancing causal event detection via corpora like Causal News. They are affiliated with SCC (Data Science) and UCREL (University Centre for Computer Corpus Research on Language), contributing to interdisciplinary projects in computational linguistics and AI ethics.
Publications span NLP applications in social media monitoring, regulatory compliance, and multilingual systems. While no awards are explicitly listed, their work reflects engagement with NLP's societal impacts and technical challenges in model transparency and resource-poor languages.
