DOI resolved by resea

Matrix Factorization Techniques for Recommender Systems

As the Netflix Prize competition has demonstrated, matrix factorization models are superior to classic nearest neighbor techniques for producing product recommendations, allowing the incorp…

Yehuda Koren, Robert Bell, Chris Volinsky
https://resea.org/10.1109/mc.2009.263

Abstract

As the Netflix Prize competition has demonstrated, matrix factorization models are superior to classic nearest neighbor techniques for producing product recommendations, allowing the incorporation of additional information such as implicit feedback, temporal effects, and confidence levels.