Harnessing sentiment analysis for political campaigns: Trends, platforms, and future directions

This reflection examines the role of social media engagement in predicting voter support through a comprehensive review of studies on sentiment analysis in political campaigns from 2010 to 2025. Synthesizing studies across diverse contexts (e.g., U.S., Europe, India, Indonesia) and platforms (e.g., Facebook, Twitter/X, TikTok, Reddit), this reflection identifies six key thematic clusters: proxy for voter sentiment, platform dynamics, technical advances, methodological and ethical issues, polarization, and campaign implications. This reflection highlights key turning points, such as the shift from Twitter-centric approaches to multi-platform analytics, the rise of deep learning, and the emergence of ethical criteria. The future work proposes a Voter Sentiment Index (VSI) as a standardized, cross-platform metric to improve interpretability and ethical application, ensuring robust and responsible use in political forecasting and governance.