Content Recommender System Based on Users Reviews

Abstract

The usage of recommender systems is widespread across many areas and has been demonstrated to be extremely important in several fields. The majority of conventional recommender systems rely on a user’s numerical rating of a consumed item to reflect that user’s opinion; however, these ratings are not always available. As a result, to make up for the absence of these evaluations, a new source of information represented by user-generated reviews is added to the recommendation process. This saves the user time from searching the Internet for movies among the thousands that are already available. The items that might be suggested to the user are described by content-based recommendation systems. It makes predictions about what content a user will enjoy based on a dataset and takes into account the qualities of the content they have already liked. The sentiment analysis field can be used to gather rich and extensive information from reviews about the entire item or a specific aspect. This publication provides a thorough introduction to assist researchers who wish to work with sentiment analysis and recommender systems. It provides background information on recommender systems, including their phases, techniques, and performance measures. After that, it talks about the idea of sentiment analysis and highlights its key components, such as level, approaches, and aspect-based sentiment analysis.