Enhancing Collaborative Filtering based Recommendations with optimized Singular Value Decomposition

In today's digital world, users are often overwhelmed by the sheer volume of books available. Recommendation systems help the users to navigate this overload by suggesting the most relevant and interesting books based on the preferences of them. Collaborative filtering (CF) is the most widely used method for online recommendations. This technique determines the user's reading interests by collecting preference or taste data from other readers. By examining the user-book rating matrix, CF identifies similar readers or books. Using the calculated similarity, it then predicts the book preferences that the user has not yet explored. This work enhances the book recommendation system by using the optimized Singular Value Decomposition (SVD). The nature inspiring algorithms like Particle Swarm Optimization (PSO) Algorithm, Genetic Algorithm (GA), Ant Colony Optimization (ACO) Algorithm and Grey Wolf (GW) Algorithm are used to refine the SVD process. After conducting several experiments, the results exposed the efficiency of Genetic Algorithm in significantly improving recommendation accuracy.