SENTIMENT ANALYSIS IN SOCIAL MEDIA: A MULTIDISCIPLINARY APPROACH USING AI AND BEHAVIORAL SCIENCE

Abstract

The exponential growth of user-generated content on social media platforms has presented unprecedented opportunities and challenges in analyzing public sentiment. This study adopts a multidisciplinary approach, integrating artificial intelligence (AI) techniques with behavioral science to enhance the accuracy and interpretability of sentiment analysis. The research applies both classical machine learning and deep learning models, augmented by psychological heuristics, to capture nuanced sentiments across various domains such as politics, health, and consumer behavior. Results show that AI models like BERT outperform traditional classifiers, especially when behavioral features are included. The findings advocate for an interdisciplinary framework to improve sentiment detection and understanding in digital spaces.