Exploring Sentiments in Text: A Survey of Implicit and Explicit Aspect-Based Sentiment Analysis

This study undertook an exploration of natural language processing (NLP), emphasizing the substantial role of sentiment analysis (SA) and aspect-based sentiment analysis (ABSA) within this domain. Given the proliferation of social media and digital interaction, the relevance of these fields has grown remarkably. The research identified a robust body of work dedicated to explicit aspect extraction, while implicit aspect extraction presented as a less developed area. Both implicit and explicit sentiment analyses were examined, recognizing the unique challenges and opportunities associated with each. Implicit sentiment analysis, despite its inherent complexity, was lauded for its capacity to interpret context without necessitating labeled data. In contrast, explicit sentiment analysis, while simpler in its implementation, was noted for its dependency on labeled data and the challenges it faces regarding generalization. The research advocated for a combined strategy to manage both aspects concurrently and laid out potential trajectories for future research. The necessity of a standard dataset for both implicit and explicit aspect extraction was stressed, and the use of recent advances in deep learning methods was advocated. In summation, the research emphasized the call for a more comprehensive approach to sentiment analysis and the pressing need for continued exploration and advancement in this area.