"anonymized dataset"

"This dataset contains anonymized social media posts related to semaglutide (Ozempic, Wegovy, and Rybelsus) collected between July 2021 and April 2024. The dataset includes textual content and metadata from X (formerly Twitter), processed to extract adverse drug reactions (ADRs) using a biomedical natural language processing pipeline. Extracted ADRs are normalized and structured for comparison with FDA-approved clinical trial data. The dataset supports research in pharmacovigilance, real-world evidence analysis, and biomedical text mining. It enables investigation of differences between patient-reported experiences and clinical trial findings across multiple levels of granularity, including brand-level and aggregated analyses, and facilitates reproducible research in drug safety monitoring."