Program Schedule

With nearly 5 billion social media users worldwide, the vast amount of data on social media presents significant opportunities and challenges for mining it as a source of health-related knowledge discovery. The Social Media Mining for Health Applications (#SMM4H) workshop provides a forum for the AMIA community to discuss health informatics at the convergence of social media, natural language processing (NLP), and machine learning. The workshop includes presentations on 5 shared task competitions that represent a variety of social media users, languages, methods, and health-related topics: (1) binary classification of English tweets self-reporting a COVID-19 diagnosis, (2) multi-class classification of sentiment associated with therapies in English tweets, (3) extraction of COVID-19 symptoms in Latin American Spanish tweets, (4) binary classification of English Reddit posts self-reporting a social anxiety disorder diagnosis, and (5) normalization of adverse drug events in English tweets. The workshop also includes a presentation on large language models for sentiment analysis, system demonstrations, a keynote speaker from academia (Luis M. Rocha, PhD, Binghamton University, “Network Science to Model Health Vulnerabilities and Biases”), a keynote speaker from industry (Raul Rodriguez-Esteban, PhD, Roche Pharmaceuticals, “Patient Listening on Social Media for Patient-Focused Drug Development”), and a panel (“What is the future of social media data for health research?”).