Assessing target audiences of digital public health campaigns: A computational approach

Abstract

As a larger proportion of society participates in social media, public health organizations are increasingly using digital campaigns to engage and educate their target audiences. Computational methods such as social network analysis and machine learning can provide social media campaigns with a rare opportunity to better understand their followers at scale. In this short paper, we demonstrate how such methods can help inform program evaluation through a case study of FDA’s The Real Cost anti-smoking Twitter campaign (@knowtherealcost). By mining publicly available Twitter data, campaigns can identify and understand key communities to help maximize reach of campaign messages to their target audiences.