A Survey on Aspect-Based Sentiment Classification

With the constantly growing number of reviews and other sentiment-bearing texts on Web, demand for automatic sentiment analysis algorithms continues to expand. Aspect-based classification (ABSC) allows extraction highly fine-grained information from text documents or sentences. In this survey, rapidly evolving state research ABSC is reviewed. A novel taxonomy proposed that categorizes models into three major categories: knowledge-based, machine learning, hybrid models. This accompanied with summarizing overviews reported model performances, both technical intuitive explanations various State-of-the-art are discussed, such as based transformer model, deep learning incorporate knowledge bases. Additionally, techniques representing inputs evaluating outputs Furthermore, trends in identified a discussion provided ways which field can be advanced future.