Data Scientist – Data Labeling and Stance Detection
As a Data Scientist at The University of North Carolina at Charlotte, I built supervised machine learning models based on semi-automated labeled datasets generated through a self-developed system and human-generated input. I also worked on stance detection of health and political misinformation on social media, as well as product review stance detection. My work involved both creating and labeling datasets for model training and evaluation. • Developed labeling workflows for credibility assessment and stance detection tasks. • Managed a self-developed labeling system integrating human input for dataset generation. • Labeled social media posts and product reviews for stance and credibility. • Ensured high-quality, accurate ground truth data for model training and evaluation.