Informatics Research Seminar: Characterizing Diet, Diabetes, Exercise, and Obesity Comments on Twitter Using Unsupervised Machine Learning
September 18 @ 4:00 pm - 5:00 pm EDT
Speaker: George Shaw Jr., PhD
Presented from UNC-C
Broadcast Link: Seminar
At Duke, all seminars live or broadcast will be held in Hock Auditorium from 4-5 pm.
Twitter provides health researchers with the opportunity to collect health related information from an unstructured data source. Previous studies have demonstrated Twitter’s ability to collect information and monitor physical activity, dietary habits, and life satisfaction. Few studies have utilized a small-scale Twitter study to retrieve user-generated content regarding diabetes, diet, exercise, and obesity (DDEO) to characterize the topics associated with them. Using sentiment analysis and unsupervised topic modeling, the works presented will describe how these two computational approaches facilitate the analytical characterization of DDEO. Highlighted will be the knowledge health practitioners can gain from the computational approaches utilized and how we move from knowledge discovery to knowledge use.
George Shaw, Jr., PhD is an Assistant Professor in the Department of Public Health Sciences at UNC Charlotte. He joined the department in August of 2018 after receiving his doctorate degree in Library and Information Science at the University of South Carolina. His current research focuses on the intersection of computational science and public health; he conducts computational text analysis on social media data to understand the connecting health topics of diet, diabetes, exercise, and obesity. Broadly, his research interests include text mining, machine learning, health literacy, information systems, and health information-seeking behaviors.