One size fits all: Enhanced zero-shot text classification for patient listening on social media

Patient-focused drug development (PFDD) represents a transformative approachthat is reshaping the pharmaceutical landscape by centering on patients throughoutthe drug development process. Recent advancements in Artificial Intelligence(AI), especially in Natural Language Processing (NLP), have enabled the analysisof vast social media datasets, also called Social Media Listening (SML), providinginsights not only into patient perspectives but also into those of other interestgroups such as caregivers. In this method study, we propose an NLP frameworkthat—given a particular disease—is designed to extract pertinent information relatedto three primary research topics: identification of interest groups, understandingof challenges, and assessing treatments and support systems. Leveraging externalresources like ontologies and employing various NLP techniques, particularlyzero-shot text classification, the presented framework yields initial meaningfulinsights into these research topics with minimal annotation effort.