A Comparative Analysis of Collaborative Filtering Algorithms for Recommendation Systems
Abstract
Recommendation Systems (RS) are essential tools for personalized information filtering. Collaborative filtering (CF) is a widely used technique in RS that leverages the collective wisdom of users to make recommendations. It analyzes user-item interaction data to identify similar users or items and generate personalized recommendations based on their behavior and preferences. In this work, we provide a comparative analysis of three prominent collaborative filtering algorithms: Matrix Factorization, Neighborhood Models, and emerging Neural Collaborative Filtering. Through our comparative analysis, we evaluate these algorithms through numerical experiments with different model structures and training parameters, and discuss their performance. By gaining a deeper understanding of these algorithms, we aim to provide valuable insights for researchers, practitioners, and businesses looking to enhance their recommendation systems and deliver more personalized experiences to their users.
Joint work with Xinyu Li.