AI-Driven Voter Sentiment and Emotion Analysis: Facebook and Youtube As Political Arenas

The rapid advancement of AI has transformed the landscape of digital sociology, offering new opportunities for understanding social phenomena, including political campaigns. This paper explores the methodological challenges of analyzing voter sentiment and emotion in the upcoming U.S. presidential election, focusing on Facebook and news-oriented YouTube channels as primary data sources. While both platforms are rich in political content, they pose distinct challenges. Facebook offers extensive user interactions through comments, likes, and shares, but its closed ecosystem and privacy policies limit data accessibility. Meanwhile, YouTube’s combination of video content and user engagement in comments presents additional complexities in analyzing multimodal data. To address these challenges, our study employs natural language processing (NLP) and machine learning algorithms to analyze voter sentiment across these platforms, with an emphasis on detecting nuanced emotions and underlying political attitudes. Using algorithms designed to process and interpret non-textual data, we extend beyond basic sentiment analysis to uncover patterns in how political messages are framed, shared, and reacted to within these different venues of digital communication.

Our findings aim to contribute to the ongoing discussion of how AI-driven methodologies can enhance the understanding of political behavior while highlighting the importance of ethical considerations and transparency. This research offers valuable insights into the evolving role of social media platforms in shaping public opinion during critical political moments.