Carine Moubarak

– Subject: Representation Learning for Healthcare Data – Performed exploratory data analysis (EDA) on complex tabular healthcare datasets to assess data quality, uncover correlations, and guide modeling strategies. – Designed and implemented state-of-the-art transformer-based models for tabular data representation learning, improving predictive performance and generalization. – Developed automated representation learning methods to replace manual feature engineering, especially in relational healthcare datasets. – Applied and evaluated models on real-world healthcare tasks, including disease risk prediction, treatment response forecasting, and personalized care strategies.

– Developed a full-stack web application for caterers and clients using ReactJS and Firebase as part of a final project.