Analysis of Movie Recommendation System Data Sets using machine learning techniques

Multimedia has emerged as one of the top entertainment source due to cheap and uninterrupted availability of high internet speeds. “Movie recommendation system have attracted much research interest within the field of recommendation systems. Two widely used techniques, one is collaborative filtering (CF) and second is content-based (CB). However, the accuracy performance of any hybrid system which utilizes more advantage of both systems to better results. Movie recommendation systems has suffered from different problems, such as “, Sparsity, Grey sheep problem, Cold start problem, Long-tail problem” etc. Basic Issues can be solved if we take the right choice on what kind of movies to ignore, what movies to suggest. The suggestions generated using approaches such as Linear Regression, Decision Trees, and Bayesian Analysis are examined in this study. Movie-Lens-1M and Movie-Lens-10M are the dataset considered. The results of this experiment suggest that Decision Tree and Linear Regression & Random Forest work well as compared to Bayesian Learning.