Scalable Collaborative Filtering Approaches for Large Recommender Systems

By Gábor Takács, István Pilászy, Bottyán Németh, Domonkos Tikk. 2009

549 citations recorded by the source corpus.

How researchers see this paper

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

In Information Systems

Citations in the corpus

Federated Collaborative Filtering for Privacy-Preserving Personalized Recommendation System (2019, background): “Low-rank factorization approaches similar to [7], [8] and [9] have been shown to handle sparsity well and are able to scale the retrieval to large datasets.”
A Comprehensive Survey of Neighborhood-based Recommendation Methods (2011, methodology): “Model-based approaches for the task of recommending items are numerous and include Bayesian Clustering [10], Latent Semantic Analysis [32], Latent Dirichlet Allocation [9], Maximum Entropy [78], Boltzmann Machines [64], Support Vector Machines [27], and Singular Value Decomposition [4] [42] [57] [74”
PowerLyra: Differentiated Graph Computation and Partitioning on Skewed Graphs (2019, methodology): “SGD [69] d=5 d=20 d=50 d=100 PowerGraph w/ Grid 15 / 35 17 / 48 21 / 73 28 / 115 PowerLyra w/ Hybrid 16 / 26 19 / 33 19 / 43 20 / 59”
A hierarchical model for ordinal matrix factorization (2011, methodology): “The addition of a local neighbour-based correction term to predictions obtained from global matrix factorization improves predictions on the Netﬂix test set; this is illustrated by Takács et al. (2009).”
Position-Transitional Particle Swarm Optimization-Incorporated Latent Factor Analysis (2022, methodology): “As illustrated by prior research [22, 25, 26], a stochastic gradient descent (SGD) algorithm is highly efficient in building an LFA model.”
Bayesian Matrix Factorization with Side Information and Dirichlet Process Mixtures (2010, methodology): “Finally, there exist numerous non-Bayesian matrix factorization techniques, including variants of singular value decomposition, and these techniques have been successfully applied to the Netflix problem (Koren 2008; Takács et al. 2009).”
An Efficient Non-Negative Matrix-Factorization-Based Approach to Collaborative Filtering for Recommender Systems (2014, methodology): “The representative works include biased SVD model [14], biased regularized incremental simultaneous MF model [15], model [3], [16], probabilistic MF model [17], and nonparametric MF model [18].”
Online Machine Learning in Big Data Streams (2018, methodology): “We can also treat users and items differently, for example, updating user vectors more dynamically than item vectors, as first suggested by [216].”
Recommender systems survey (2013, background): “Matrix factorization is especially adequate for processing large RS databases and providing scalable approaches [215].”
Location recommendation for location-based social networks (2010, background): “Hence, scalability is a critical issue since large-scale matrix multiplication incurs tremendous computational overhead for collaborative recommendation [ 5 , 6, 7].”
The use of machine learning algorithms in recommender systems: A systematic review (2015, unknown): “Movie [5] [5] [5] [14] [36] [39] [53] [53] [60] [60] 10”