Evaluating the Effects of Misinformation on Public Sentiments Surrounding Access to Abortion Through Social Media Sentiment Analytics

Abstract

As social media use has grown in recent years, ease of access and rapid data collection through online social media has permitted researchers to measure and track sentiments related to emerging public health threats. Herein, we explore the possibilities of examining messaging shared via social media networks for sentiment classification as it relates to women's reproductive healthcare, especially access to abortion. In our previous works, our team has successfully employed various natural language processing (NLP) models for the analysis of social media shared sentiments. This study reports a work-in-progress on the similar use of fine-tuned NLPs (i.e., DistilRoBERTa) to collect/analyze the sentiments of socio-behavioral data shared via social networks to uncover a correlation between reproductive-related misinformation (i.e., access to abortion) and public sentiments/discourse direction.