Recommender System Based On Alternating Least Squares Collaborative Filtering Algorithm

In the current era,Web is the best source for getting any information or making a decision on something.People get online suggestions before making any decision such as buying any product,booking movie tickets etc.In such cases,recommender systems play an important role.Recommender systems work on the data about users and items which has to be recommended.Collaborative filtering is one of the well known and most extensive techniques of a recommender system.The basic idea is to predict which items a user would be interested in based on their preferences.Recommender systems using collaborative filtering are able to provide an accurate prediction when enough data is provided because this technique is based on the user's preference.Many recommender systems suggest items to users employing collaborative filtering techniques,which process historical records of items that the users have viewed,purchased,or rated.Two major problems that most collaborative filtering approaches have to resolve are scalability and sparseness of the user's profile matrix,which has been successfully overcome with the use of latent factor models technique.The most successful realizations of latent factor models are based on matrix factorization.Among the algorithms for matrix factorization,Alternating Least Squares(ALS)stands out because its computations are easily parallelizable.In this work,we propose a methodology for implementing a Movie Recommender System by using the sate of art algorithm executed in Apache Spark framework.We perform experiments to evaluate the accuracy of generated recommendations and the execution time of the the algorithm,using publicly available MovieLens dataset.Experimental results show that running the Alternating Least Squares algorithm on Spark framework is fast and more efficient.