Abstract
Diabetes is one of the most common illnesses in the U.S. and causes serious damage to daily life. Patients and their significant others are exposed to a variety of messages encouraging healthy behaviors. However, it is unclear as to what specific messages are most effective in discussing the issue. The study is designed to gain insights into consumer preferences for information about diabetes by analyzing content on the online question and answer platform, Quora. It investigates these preferences through the perspective of persuasive communication emphasizing Aristotle’s Rhetoric including ethos (source credibility), pathos (emotional appeal), and logos (logical reasoning). The approach includes a comparative analysis of the most popular answers with answers with fewer upvotes on the same questions so as to examine how the two different kinds of answers differ in applying rhetorical principles. A total of 200 pairs of the most upvoted and less upvoted answers will be collected. Utilizing a fine-tuned large language model (LLM), ChatGPT, this study will highlight the distinguishing features of the most upvoted answers compared to the less popular ones. This comparison will elucidate which persuasive elements align with consumer preferences regarding diabetes information. This research contributes to the field of health information by identifying the types of messages that resonate with consumers, thereby potentially increasing the likelihood of their acceptance and application of health information. Moreover, it demonstrates the practicality of employing an LLM for the content analysis of health information, particularly in uncovering latent content, which extends AI-driven text analysis.
Presenters
Beom BaeAssociate Professor, Communication Arts, Georgia Southern University, Georgia, United States Sang Won Bae
Student, Bachelors, University of Georgia, Georgia, United States
Details
Presentation Type
Paper Presentation in a Themed Session
Theme
KEYWORDS
Diabetes, Persuasion, Large language model, Health information, Artificial intelligence
Digital Media
This presenter hasn’t added media.
Request media and follow this presentation.
