23. SOCIAL MEDIA ANALYTICS FOR UNDERSTANDING PUBLIC SENTIMENT: A DATA SCIENCE PERSPECTIVE

In the digital era, social media platforms such as Twitter, Facebook, Instagram, and TikTok have become central arenas for individuals to express opinions and emotions. These platforms generate vast volumes of textual, visual, and multimedia data that reflect collective sentiment in near real-time. Sentiment analysis, an interdisciplinary area combining natural language processing (NLP), machine learning, and data science, has emerged as a critical tool for analyzing this data. This paper provides a systematic review of foundational and recent contributions in social media sentiment analysis, with emphasis on methodological advances such as deep learning, transformer-based models, multilingual approaches, and multimodal fusion. Applications across politics, marketing, crisis management, and public health are discussed, alongside key challenges including misinformation, bias, and ethical considerations. The review highlights future research opportunities in explainability, low-resource language modeling, and ethical frameworks, demonstrating the potential of social media analytics to inform decision-making across disciplines.