A hybrid approach for a movie recommender system using Content-Based, Collaborative and Knowledge-Based Filtering methods

Rapid evolution of information technology has rapidly increased the significance of Internet-generated data. As a result, finding specific information within this vast data landscape has become increasingly challenging for individuals. Recommender systems have emerged as precious tools in response to this challenge, offering assistance in navigating and extracting relevant information from the abundance of available data. As user engagement increases in importance, recommender systems will be considered an integral part of personalized marketing. The most common methods leverage product features (Content-Based), user similarity (Collaborative Filtering), and personal information (Knowledge-based). However, with the increasing popularity of Neural Networks, companies have started experimenting with new Hybrid Recommendation Systems that combine them all. This study thoroughly assesses the strengths and limitations of prevalent filtering methods, aiming to introduce an advanced hybrid recommender system. By integrating content-based, collaborative, and knowledge-based techniques, the hybrid model notably enhances recommendation accuracy, effectively addressing challenges like the "cold start" problem and data sparsity issues.