Improved Content Based Filtering Using Unsupervised Machine Learning on Movie Recommendation

Abstract

In our world of massive entertainment options and with thousands of choices on every movie platform, the user found himself in the circle of confusion over which movie to choose. Here the solution is using the recommender systems to predict user’s interests and recommend items most likely to interest them. Recommender systems are utilized in a variety of areas and are most commonly recognized as playlist generators for video and music services, product recommenders for online stores as AliExpress and Amazon..., or content recommenders for social media platforms and open web content recommenders.

In this paper, we propose a new powerful recommender system that combines Content Based Filtering (CBF) with the popular unsupervised machine learning algorithm K-means clustering. To recommend items to an active user, K-means is then applied to the movie data to give each movie a specific cluster and after founding the cluster to which the user belongs, the content-based approach applies to all movies with the same cluster. The experimentation of well-known movies, we show that the proposed system satisfies the predictability of the Content-Based algorithm in GroupLens. In addition, our proposed system improves the performance and temporal response speed of the traditional collaborative filtering technique and the content-based technique.