Sentiment Analysis in Online Public Opinion: Trends, Challenges, and Future Directions

Abstract

The rapid development of the Internet, particularly following the COVID-19 pandemic, has significantly influenced public opinion analysis, a critical field in natural language processing (NLP). This paper systematically reviews recent domestic and international research on online public opinion from an emotional perspective, drawing on sources from the China National Knowledge Infrastructure (CNKI) and Web of Science. The study highlights the importance of sentiment analysis in understanding public opinion, emphasizing the need for advanced machine learning techniques and multimodal content integration to enhance model performance. Future research should explore the inclusion of diverse languages and larger datasets to improve generalization, and consider transforming sentiment detection tasks into multi-classification or regression challenges. This comprehensive review aims to guide future researchers by providing a deeper understanding of current trends and potential advancements in online public opinion analysis.