Movie Recommendation Using K-Nearest Neighbors Algorithm

Abstract

In a landscape brimming with an extensive array of movies globally, navigating through the multitude of options to find films that suit one's unique preferences can prove to be a difficult assignment for viewers. The sheer volume of choices often leaves individuals feeling overwhelmed, presenting a challenge in selecting a movie that resonates with their tastes. Consequently, movie Service providers are accountable for the duty of furnishing a recommendation system that enhances the user experience by assisting them in discovering movies that complement their preferences. Previous research into recommendation systems, particularly those employing Machine Learning (ML) algorithms, has proved they were better than the conventional recommendation methods. However, there remains a requirement for further refinement, especially in circumstances in which users face difficulties in identifying movies within their favorite genres. Prolonged looks for the appropriate film can exacerbate issues such as information scarcity and chilly launch challenges. To deal with these problems effectively, we propose the implementation of a recommender system based on machine learning focused on movie genres, leveraging the algorithm known as KNN, or K-nearest Neighbors. Our proposed solution features a user-friendly interface hosted in a well-lit web application, equipped utilizing a slider bar functionality that empowers users to specify their taste in movies and get personalized suggestions for similar titles. Through the integration of user feedback and choices, our system aims to deliver customized advice that is better suited to individual interests and tastes, thereby improving the overall movie-viewing experience.