Fusion of Various Sentiment Analysis Techniques for an Effective Contextual Recommender System
Ananth Gouri S, Raghuveer, S. Vasanth Kumar · 2023
In recent years, Recommendation Systems (RSs) have become a vital component of numerous websites and online applications in a variety of domains. Consider, for instance, the e-commerce websites where RSs predominate. In part, the problem of information overload is mitigated by these RSs. However, RSs still have a few issues, such as data sparsity, which leads to another issue known as cold-start. The cold-start problem occurs when the user-item matrix is incredibly sparse. Additionally, RSs has a problem known as the long-tail problem, in which the system is incapable of providing suggestions due to insufficient or invalid ratings for often purchased products. The cold-start problem can be solved by providing recommendations based on captured user preferences and user feedback. User attitudes can be gleaned from the analysis of textual user reviews of purchased products. Sentiment analysis(SA), also known as opinion mining, is the study of people’s opinions, feelings, judgments, feedbacks, and emotions conveyed through written language regarding entities and their features. Usually these sentiments are derived and based out of various contexts. A context in sentiment analysis is a mood-based natural attribute.