Using machine learning techniques to investigate learner engagement with TikTok media literacy campaigns

Abstract Social media has the unique capacity to expose many learners to media literacy instruction via targeted campaigns. Investigating learner engagement and reaction to these efforts may be a fruitful endeavor for researchers that can inform the design of future campaigns. However, the massive datasets associated with social media posts are difficult, and often impossible, to analyze with traditional qualitative methods. This study seeks to address this problem by leveraging machine learning techniques to collect and analyze Big Data from two different media literacy campaigns on the youth-oriented social media platform TikTok. Specifically, we explore the ways topic modeling, sentiment analysis, and network analysis can provide insight into learner engagement with these campaigns and discuss limitations and implications for stakeholders interested in utilizing these approaches.