Comparative Review on Sentiment analysis-based Recommendation system

Recommendation systems are ubiquitous these days and are used in nearly every domain; from learning which videos could be recommended to users on streaming websites, to products that can be sold on e-commerce platforms. These systems are driven by the copious amount of data that is scraped and collected from sources such as review platforms and social media websites. On this collected data, sentiment analysis can be performed to recommend products to users based on an overall analysis of sentiments conveyed using reviews, comments, or opinions. The information thus obtained is provided to already existing machine learning-based filtering techniques which include content-based, collaborative, and hybrid filtering. The aim of this paper is to provide a detailed review of various techniques used for sentiment-based recommendation systems and the inherent challenges in these techniques.