Funded Projects

Social big data analysis of publicly available user data on social media platforms is a promising approach for attaining organic observations of behavior that can monitor and predict real-world public health problems, such as HIV incidence. In preliminary research, our team identified and collected tweets suggesting HIV risk behaviors (e.g., drug use, high-risk sexual behaviors), modeled them alongside CDC statistics on HIV diagnoses, and found a significant positive relationship between HIV-related tweets and county-level HIV cases. We propose to create a single automated platform that collects social media data, identifies and labels tweets that suggest HIV-related behaviors, and predicts regional HIV incidence. We will interview staff and participants at local and regional HIV organizations to understand ethical issues associated with mining people’s data. The software developed from this application will be shared with HIV researchers and health care workers to combat the spread of HIV.