Michael Paul: Analyzing Reactions to a Disease Epidemic: An NLP Approach
This talk will present ongoing research studying a large corpus of Twitter messages related to the 2015 Zika virus outbreak in North and South America using natural language processing (NLP). I will first present a topic analysis of the corpus, highlighting how the themes of discussion vary by time and location. Second, I will present work on measuring a particular behavioral response to the epidemic: changing travel plans in response to travel advisories. For both parts, I will describe improvements to NLP methods to support the applications.
Michael Paul is an Assistant Professor of Information Science at the University of Colorado Boulder. His research combines machine learning and natural language processing with applications in social media analysis and health informatics. He recently co-authored a book, Social Monitoring for Public Health. He received a PhD in Computer Science from Johns Hopkins University in 2015, and a BS in Computer Science from the University of Illinois at Urbana-Champaign in 2009.
The Information Science seminar is a weekly talk series and gathering for the Information Science department and its extended community. Any faculty, students, and interested parties regardless of affiliation are welcome. The seminar will meet most weeks during Fall 2018 on Wednesdays at 1pm. Keep an eye out for future announcements!
Wednesday, September 5, 2018 at 1:00pm to 2:00pm
Center for Academic Success, W313 1725 Euclid Avenue, Boulder