Movie Recommender System Using Collaborative Filtering
Meenu Gupta, Aditya Thakkar, Aashish, Vishal Gupta, Dhruv Pratap Singh Rathore
Abstract
Movies are one of the sources of entertainment, but the problem is in finding the desired content from the ever-increasing millions of content every year. However, recommendation systems come much handier in these situations. The aim of this paper is to improve the accuracy and performance of a regular filtering technique. Although varieties of methods are used to implement a recommendation system, Content-based filtering is the simplest method. Which takes input from the users, rechecks his/her history/past behavior, and recommends a list of similar movies. In this paper, to prove the effectiveness, K-NN algorithms and collaborative filtering are used to mainly focus on enhancing the accuracy of results as compared to content-based filtering. This approach is based on cosine similarity using k-nearest neighbor with the help of a collaborative filtering technique, at the same time removing the drawbacks of the content-based filtering. Although using Euclidean distance is preferred, cosine similarity is used as the accuracy of cosine angle and the equidistance of movies remain almost the same.