Matrix Factorization Techniques for Recommender Systems
Yehuda Koren, Robert Bell, Chris T. Volinsky · Computer · 2009
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.
Cited by 11,673 works. doi:10.1109/mc.2009.263
Authors
References
- Using collaborative filtering to weave an information tapestry
- Factorization meets the neighborhood
- Probabilistic Matrix Factorization
- Application of Dimensionality Reduction in Recommender System - A Case Study
- The Netflix Prize
- Collaborative Filtering for Implicit Feedback Datasets
- Scalable Collaborative Filtering with Jointly Derived Neighborhood Interpolation Weights
- Collaborative filtering with temporal dynamics
- Large-Scale Parallel Collaborative Filtering for the Netflix Prize
- Major components of the gravity recommendation system