Web-Based Movie Recommender System

Abstract

Recommender systems guide users to find and cull items such as restaurants, books, and movies from the huge collection of available options on the Web. Based on the user’s taste and preferences, recommender system recommends a set of items from a large set of available items. Recommendation systems include intelligent aides for filtering and selecting Web sites, news stories, TV listings, and alternative info. The users of such systems typically have various conflicting desires, variations in personal preferences, social and academic backgrounds, and personal or skilled interests. As a result, it looks fascinating to possess personalized intelligent systems that suit individual preferences. The need for personalization has proliferated the event of systems that adapt themselves by ever-changing their behavior supported by the inferred characteristics of the user interacting with them. In this paper, we develop a Web-based movie recommender system that recommends movies to users based on their profile using different recommendation algorithms. We also compare various recommendation algorithms such as singular value decomposition, alternating least squares and restricted Boltzmann machines.