Analyzing and improving stability of matrix factorization for recommender systems

dc.contributor.authorD'Amico, Edoardo
dc.contributor.authorGabbolini, Giovanni
dc.contributor.authorBernardis, Cesare
dc.contributor.authorCremonesi, Paolo
dc.contributor.funderScience Foundation Irelanden
dc.contributor.funderEuropean Regional Development Funden
dc.description.abstractThanks to their flexibility and scalability, collaborative embedding-based models are widely employed for the top-N recommendation task. Their goal is to jointly represent users and items in a common low-dimensional embedding space where users are represented close to items for which they expressed a positive preference. The training procedure of these techniques is influenced by several sources of randomness, that can have a strong impact on the embeddings learned by the models. In this paper we analyze this impact on Matrix Factorization (MF). In particular, we focus on the effects of training the same model on the same data, but with different initial values for the latent representations of users and items. We perform several experiments employing three well known MF implementations over five datasets. We show that different random initializations lead the same MF technique to generate very different latent representations and recommendation lists. We refer to these inconsistencies as instability of representations and instability of recommendations, respectively. We report that stability of item representations is positively correlated to the accuracy of the model. We show that the stability issues affect also the items for which the recommender correctly predicts positive preferences. Moreover, we highlight that the effect is stronger for less popular items. To overcome these drawbacks, we present a generalization of MF called Nearest Neighbors Matrix Factorization (NNMF). The new framework learns the embedding of each user and item as a weighted linear combination of the representations of the respective nearest neighbors. This strategy has the effect to propagate the information about items and users also to their neighbors and allows the embeddings of users and items with few interactions to be supported by a higher amount of information. To empirically demonstrate the advantages of the new framework, we provide a detailed description of the NNMF variants of three common MF techniques. We show that NNMF models, compared to their MF counterparts, largely improve the stability of both representations and recommendations, obtain a higher and more stable accuracy performance, especially on long-tail items, and reach convergence in a fraction of epochs.en
dc.description.sponsorshipScience Foundation Ireland (SFI under Grant Number 12/RC/2289-P2, which is co-funded under the European Regional Development Fund)en
dc.description.statusPeer revieweden
dc.description.versionAccepted Versionen
dc.identifier.citationD'Amico, E., Gabbolini, G., Bernardis, C. and Cremonesi, P. (2022) 'Analyzing and improving stability of matrix factorization for recommender systems', Journal of Intelligent Information Systems, doi: 10.1007/s10844-021-00686-1en
dc.identifier.journaltitleJournal of Intelligent Information Systemsen
dc.relation.projectinfo:eu-repo/grantAgreement/SFI/SFI Research Centres/12/RC/2289/IE/INSIGHT - Irelands Big Data and Analytics Research Centre/en
dc.rights© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2021. This version of the article has been accepted for publication, after peer review (when applicable) and is subject to Springer Nature’s AM terms of use, but is not the Version of Record and does not reflect post-acceptance improvements, or any corrections. The Version of Record is available online at:
dc.subjectMatrix factorizationen
dc.subjectNearest neighborsen
dc.subjectPopularity biasen
dc.titleAnalyzing and improving stability of matrix factorization for recommender systemsen
dc.typeArticle (peer-reviewed)en
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
789.33 KB
Adobe Portable Document Format
Accepted version
License bundle
Now showing 1 - 1 of 1
Thumbnail Image
2.71 KB
Item-specific license agreed upon to submission