Advances in Collaborative Filtering

By Yehuda Koren, Robert M. Bell. 2011

1,517 citations recorded by the source corpus.

Subfields: Computational Theory and Mathematics, Information Systems, Management Science and Operations Research, Recommender Systems

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Theory and Mathematics

In Information Systems

In Management Science and Operations Research

In Recommender Systems

Citations in the corpus

Ups and Downs: Modeling the Visual Evolution of Fashion Trends with One-Class Collaborative Filtering (2016, methodology): “We begin by briefly describing the underlying ‘standard’ Matrix Factorization method [20], whose basic formulation we adopt.”
VBPR: Visual Bayesian Personalized Ranking from Implicit Feedback (2015, methodology): “Our preference predictor is built on top of Matrix Factorization (MF), which is state-of-the-art for rating prediction as well as modeling implicit feedback, whose basic formulation assumes the following model to predict the preference of a user u toward an item i (Koren and Bell 2011):”
Image-Based Recommendations on Styles and Substitutes (2015, background): “The system shares these characteristics with more common forms of recommender system, but does so on the basis of the appearance of the object, rather than metadata, reviews, or similar.”
From amateurs to connoisseurs: modeling the evolution of user expertise through online reviews (2013, neutral): “lf: A standard latent-factor recommender system [21].”
StarSpace: Embed All The Things! (2017, methodology): “2001) and its improvements such as SVD++ (Koren and Bell 2015), as well as a host of other techniques, e.”
Hidden factors and hidden topics: understanding rating dimensions with review text (2013, methodology): “1), for instance alternating least-squares, and gradient-based methods [11].”
Fusing Similarity Models with Markov Chains for Sparse Sequential Recommendation (2016, background): “In spite of not explicitly parameterizing each user, FISM surprisingly outperforms various competing baselines, including Matrix Factorization, especially on sparse datasets [2].”
Protecting privacy when mining and sharing user data (2012, methodology): “Furthermore, the contest has given high publicity and led to significant progress in the field of recommender systems [23,107], resulting in establishment of matrix factorization techniques as the dominant methodology for implementing collaborative filtering [108, 177], introduction of two-layer und”
Graph Neural Networks in Recommender Systems: A Survey (2020, methodology): “Broadly speaking, in the past decades, the mainstream modeling paradigm in recommender systems has evolved from neighborhood methods [6, 60, 95, 123] to representation-learning-based frameworks [25, 77, 78, 125, 143].”
Recommender Systems (2016, positive): “Such models are suitable in settings in which the item space changes rapidly, but the users are relatively stable over time [312].”
Evaluating Prerequisite Qualities for Learning End-to-End Dialog Systems (2015, background): “Singular Value Decomposition (SVD) is a standard benchmark for recommendation, being at the core of the best ensemble results in the Netﬂix challenge, see Koren & Bell (2011) for a review.”
Quantum Recommendation Systems (2016, background): “There has also been a series of works in the machine learning community many of them inspired by a practical challenge by Netflix on real world data [14, 6, 13].”
Session-based Recommendation with Graph Neural Networks (2018, background): “Matrix factorization (Mnih and Salakhutdinov 2007; Koren, Bell, and Volinsky 2009; Koren and Bell 2011) is a general approach to recommendation systems.”
Differentially Private Matrix Completion, Revisited (2017, background): “Collaborative filtering (or matrix completion) is a popular approach for modeling the recommendation system problem, where the goal is to provide personalized recommendations about certain items to a user [32].”
An Efficient Non-Negative Matrix-Factorization-Based Approach to Collaborative Filtering for Recommender Systems (2014, methodology): “On the other hand, Tikhonov regularization has proved to be highly reliable in improving the accuracy of MF-based CF models [3], [13]–[18]; therefore, it is quite meaningful and significant to integrate this regularization strategy into the NMF-based CF model, which has been never seen in any prior…”
Self-Attentive Sequential Recommendation (2018, background): “dimensions to represent users’ preferences and items’ properties, and estimate interactions through the inner product between the user and item embeddings [6], [7].”
Neural Collaborative Filtering vs. Matrix Factorization Revisited (2020, background): “This modification does not add expressiveness but has been found to be useful in many studies, likely because its inductive bias is better suited to the problem [23, 32].”
Representation Learning with Large Language Models for Recommendation (2023, background): “LLMs have the potential to generate text answers that may include recommendations for non-existent items due to hallucination issues [20].”
Latent factor models with additive and hierarchically-smoothed user preferences (2013, methodology): “One of the most extensively used techniques in these recommendation systems is the latent factor model and its variants [2] [13] [14] [16] (see Koren et al. [15] for an excellent survey).”