Generating Book Suggestions Using a Hybrid Model with Sentiment Analysis

Recommendation Systems are ubiquitous, whether it is in social media posts, advertisements, or digital book and movie libraries. Many recommender systems suggest items based on the product or its use and do not consider the user's sentiments towards the item or user preferences. This paper employs a hybrid model comprising a mix of popularity and content-based approaches to generate book suggestions along with sentiment analysis. Sentiment analysis of the book descriptions and user feedback enhances the recommendations. This recommendation system also incorporates classification and filtering of explicit books based on the user's age to generate the top 10 recommendations of books. The measurement metrics implemented in testing this proposed model are Precision, Recall, Mean Absolute Error, and Mean Squared Error to test the recommendation system's performance. The proposed system with high precision and low error results in notable additions to existing recommendation systems.