Analyzing Social Media Data to Identify Trends in Use of Tobacco-Related Products

Almost 500,000 Americans die each year from tobacco-related illnesses, such cancer and heart disease. To improve Food and Drug Administration (FDA) surveillance of social media trends related to tobacco products, AIR built two social media analysis tools that leverage machine learning, text analytics, and natural language processing to scrape and analyze content on social network websites.

The first tool extracts X (formerly Twitter) social media data and conducts social network analyses to help track topic trends and influencers and detect social network communities. The second tool conducts content analyses using natural language processing and topic modeling on the extracted data to better understand the content of the posts and their impacts. Together, these tools can be used, for example, to assess youth perceptions and industry advertising tactics related to vaping and other electronic nicotine delivery system products.