Movie Recommendation Using Clustering and Nearest Neighbour

Abstract

Due to the abundance of items available and online information, a user cannot easily choose which product is ideal for him. A recommender system assists users in finding what is best for them. A recommender system uses information about a user's activity. It uses it to suggest movies to users based on their individual interests. This paper provides an overview of a recommender system that uses the K-means and KNN algorithm. Without wasting time exploring, both algorithms rapidly and effectively recommend movies to customers based on their likes. There are many uses for recommender systems worldwide. K-means algorithm is used to get beyond some of the restrictions of content-based and collaborative work. The K-means algorithm creates clusters of individuals with similar interests, and KNN, which includes nearest neighbors, recommends movies to each group. This is used in well-known fields like books, news, music, videos, and movies, among others. These search engines allow users to find movies of their choice. K-mean, KNN, and hybrid algorithms have been covered in this study. K-means algorithm results based on metrics like "average Genre Rating" and "User Movie Rating". The RMSE feature has been used to KNN algorithm. A hybrid algorithm combines the two algorithms mentioned above. K-means is given an input, and the output of this method serves as the input for the KNN algorithm, which is more accurate than both K-means and KNN.