Natural vs. Artificially Sweet Tweets: Characterizing Discussions of Non-nutritive Sweeteners on Twitter

Abstract

This ongoing project aims to use social media data to study consumer behaviors regarding natural and artificial sweeteners Following the recent shifts to natural sweeteners such as Stevia versus artificial, and traditionally-used ones like aspartame in recent years, there has been discussion around potential negative side effects, including memory loss and other chronic illnesses. These issues are discussed on Twitter, and we hypothesize that Twitter may provide insights into how people make nutritional decisions about the safety of sweeteners given the inconclusive science surrounding the topic, how factors such as risk and consumer attitude are interrelated, and how information and misinformation about food safety is shared on social media. As an initial step, we describe a new dataset containing 308,738 de-duplicated English-language tweets spanning multiple years. We conduct a topic model analysis and characterize tweet volumes over time, showing a diversity of sweetener-related content and discussion. Our findings suggest a variety of research questions that these data may support.