Matrix Factorization Based Normalization in Movie Recommendation Systems

The Recommender System (RS) has grown exponentially in the last several years in many different groups. Researchers are interested in it since many companies operating on online video-sharing platforms like Netflix, YouTube, Hotstar, etc., have recently expanded. A movie recommendation system recommends movies to the viewer that will be most likely to be viewed by that user. A tailored suggestion engine may be quite helpful, particularly if the consumer doesn't know what movie they want to view. This research has suggested a collaborative filtering movie recommendation system prototype using normalization and matrix factorization techniques. Here we have decomposed the user rating matrix into two matrices with lesser dimensions. After the above process, we apply the normalization techniques on both matrices to get missing ratings. Further, both matrices are converted into two Boolean matrices, which are multiplied to get a resultant matrix. Our proposed recommendation algorithm tries to recommend the best possible movies that will be liked by the users.