Shelby Hall Graduate Research Forum Posters

Download

Download Full Poster (449 KB)

Description

The spread of medical misinformation poses significant threats to public health, healthcare system stability, and the quality of patient care. Our research examines misinformation targeting the U.S. healthcare system. It uses a mixed-methods approach that includes social media data analysis, surveys of practicing nurses, and agent-based simulation. This poster focuses on the initial phase, which attempts to detect potential misinformation and disinformation by analyzing patterns in social media posts and user account behaviors. A detailed account of the data preparation process lays the groundwork for examining how disinformation operates online. This phase draws on the Pushshift repository, which offers historical Reddit content for social media analysis. A total of 72 individual forums, called subreddits, were initially included in this study. The novel coronavirus outbreak of 2019 (COVID-19) was selected as the focus of this research, with all subreddits chosen due to their topic of discussion relating to this pandemic specifically or the nursing profession more generally. Exploratory analysis in Tableau was used to view data trends and identify preliminary patterns. However, the inherent noise, inconsistency, and unstructured nature of social media data necessitate extensive data cleaning and preprocessing. Following cleanup and preparation, the analysis looks for indicators of misinformation such as posting frequency and amplification patterns. Sentiment analysis is also used to examine the emotional tone of these expressed viewpoints. These traits will be used to determine how disinformation narratives take shape and spread online.

Publication Date

3-2026

Department

Computer Science

Disciplines

Computer Sciences

Deconstructing Digital Disinformation: Social media Data Preparation and Analysis for Healthcare Research

Share

COinS