A New Approach for Movie Recommender System using K-means Clustering and PCA

Abstract

Recommendation systems are refining mechanism to envisage the ratings for items and users, to recommend likes mainly from the big data. Our proposed recommendation system gives a mechanism to users to classify with the same interest. This recommender system becomes core to recommend the e-commerce and various websites applications based on similar likes. This central idea of our work is to develop movie recommender system with the help of clustering using K-means clustering technique and data pre-processing using Principal Component Analysis (PCA). In this proposed work, new recommendation technique has been presented using K-means clustering, PCA and sampling with the help of MovieLens dataset. Our proposed method and its subsequent results have been discussed and collation with other existing methods using evaluation metrics like Dunn Index, average similarity and computational time has been also explained and prove that our technique is best among other techniques. The results achieve from the MovieLens dataset is able to prove high efficiency and accuracy of our proposed work. Our proposed method is able to achieve the MAE of 0.67, which is better than other methods.