Movie Recommendation using Collaborative Filtering to Make Accurate Predictions
Shreya Jain, Richard Essah · 2022
A recommendation system that suggests appropriate movies to customers based on their prior interactions with the platform will increase customer happiness and, as a result, increase revenue for online streaming services like Netflix. The methods we'll cover here can be used for any product for which you wish to create a recommendation system; they are not just restricted to movies. In this paper, a collaborative filtering framework is used to provide recommendations to the users. We have used three different algorithms on the rating dataset. These are cosine similarity, K nearest neighbors (KNN), and Singular Value Decomposition (SVD) to predict the outcomes. The results are obtained in the form of Root Mean Square Error (RMSE), precision, and recall of all algorithms and at the end, we have done a comparison between these algorithms to check which algorithm is better. Many e-commerce websites like Amazon, Flipkart, and Mantra use this technique to know their customer behaviour and provide recommendations to them about the products that they are most likely to buy.