Roe v Wade in Twitter: Sentiment Analysis with Machine Learning

Abortion policy has long been a contentious issue in the United States, and the recent overturning of the Roe v Wade ruling has reignited the national debate on this topic. With the widespread use of social media, it is now easier than ever to gather data on public perceptions and attitudes towards healthcare policies such as abortion. The objective of this study is to conduct sentiment analysis on US abortion policy tweets using machine learning techniques. To achieve this goal, a random selection of tweets available to the public was gathered from January 2021 to October 2022. The tweets were filtered based on relevant hashtags related to abortion and the Roe v Wade ruling. The collected dataset comprised over 450,000 tweets, which were then analyzed using an automated sentiment scoring technique based on a lexicon-based approach. A 78.6% accuracy in predicting the sentiment of tweets was achieved through the use of BERT and deep learning. The insights gained from this study can be useful for policymakers and stakeholders in the healthcare industry to understand public sentiment towards abortion policy and formulate effective communication strategies.