Principal Investigator: Mark Dredze (CTP contact: Mario Navarro)
Funding Mechanism: Centers of Excellence in Regulatory Science and Innovation Grant (CERSI)
ID number: 3U01FD005942-04S1
Award Date: 9/1/2019
Institution: Johns Hopkins University
This research involves two projects that will use artificial intelligence (Al) methods to analyze social media posts and comments. Researchers will apply Al models to messages collected from Twitter and Reddit; these models will enable the timely identification, organization, and analysis of millions of social media posts and comments. Separate models will be developed for Twitter and Reddit, which will allow researchers to compare and contrast findings from the two social media platforms. Project 1 will detect and identify tobacco brands and products from a list of previously established brands and products. Project 2 will identify the mentions of tobacco product-related adverse events (e.g., burns from e-cigarettes, vaping lung illnesses) and the perceived health benefits and harms of tobacco products. Project findings will provide new information that may inform FDA regulatory activities and communications.