Application domains of aspect and sentiment classification techniques: A survey

Recently, social media has changed the way, information is produced, transferred and consumed. User-generated content in the form of posts, blogs, comments, feedback, and reviews, has established a new connection between the producers and users of information. Tracking such content has enabled businesses to collect customer’s feedback to provide better services. The abundance of information from diverse sources helps users tap into the wisdom of crowds and aid in making more informed decisions. This raised the question of overcoming information overload and providing a rich and coherent user experience. This question has opened a rich venue for researchers on how to analyze such a huge amount of customer feedback to get actionable insights. A lot of research has been conducted in this area, which largely depends on opinion mining, sentiment analysis and text mining algorithms to interpret and make sense of large amounts of textual data. The proposed survey aims to identify application domains. Exploring each application domain to determine which one is more complex than others. Evaluating state-of-the-art NLP and machine learning techniques against application domains. Setting a future direction by deliberating the complexity of the dataset, lack of benchmark dataset, performance comparison, and future work. Moreover, this study critically analyzed surveys on document-level sentiment analysis and aspect-based sentiment analysis surveys. In this way, the last ten years' surveys (document and aspect level) have been investigated to find out what areas these surveys are not addressing. The main focus is on aspect-based sentiment analysis techniques (ABSTs). On the other hand, this study provides such a novelty by identifying limitations in machine learning techniques, the complexity of the dataset, and identifying application domains and NLP challenges. In addition, six application domains and four limitations have been identified by exploring the recent ten years of aspect-based sentiment analysis techniques. Finally, it thoroughly explores the application domains and limitations of aspect-based sentiment analysis techniques.