L21RSVD: Robust L21 Norm SVD-Type Latent Factor Models for Rating Prediction
Algorithms · 20.07.2026 · 19 (7) · 600
DOI: 10.3390/a19070600 ↗Kurz gefasst
Fachartikel in der Zeitschrift Algorithms, online erschienen am 20.07.2026. Verfasst von Chenggang He und 2 weiteren Autor:innen. Die Arbeit nennt 37 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Opening Foundation of State Key Laboratory of Cognitive Intelligence iFLYTEK, Anhui Postdoctoral Scientific Research Program Foundation, Scientific Research Foundation for High-level Talents of Anhui University of Science and Technology.
Abstract (Originalsprache)
Recommender systems face fundamental challenges, including extreme data sparsity and noisy rating observations. We propose L21RSVD, a robust latent factor model that employs L21 norm regularization within a Singular Value Decomposition framework to mitigate the impact of outliers while preserving low-rank structure. Unlike conventional L2-regularized approaches, our formulation induces group sparsity in the latent factor space, yielding more discriminative user and item representations. We derive three optimization variants: standard L21RSVD, L21RSVD without the squared term, and coefficient-free adaptive L21RSVD. Building upon these, we introduce a fusion strategy that adaptively aggregates predictions based on local data density. Extensive experiments on benchmark datasets demonstrate that L21RSVD substantially outperforms classical collaborative filtering and the SVD-type model. The proposed fusion model achieves state-of-the-art performance, reducing RMSE by up to 31.21% and MAE by up to 42.99% relative to baseline methods.
Förderung
- Opening Foundation of State Key Laboratory of Cognitive Intelligence iFLYTEK
- Anhui Postdoctoral Scientific Research Program Foundation
- Scientific Research Foundation for High-level Talents of Anhui University of Science and Technology
- Anhui Quality Engineering Project
- Anhui University of Technology Postgraduate Education Teaching Research Project
- Natural Science Foundation of China
- NSFC Key Project of International (Regional) Cooperation and Exchanges
Zitieren
APA
He, C., Hu, C., & Qian, D. (2026). L21RSVD: Robust L21 Norm SVD-Type Latent Factor Models for Rating Prediction. Algorithms, 19(7), 600. https://doi.org/10.3390/a19070600
Vancouver
He C, Hu C, Qian D. L21RSVD: Robust L21 Norm SVD-Type Latent Factor Models for Rating Prediction. Algorithms. 2026;19(7):600. doi:10.3390/a19070600
BibTeX
@article{He2026,
author = {He, Chenggang and Hu, Can and Qian, Demeng},
title = {L21RSVD: Robust L21 Norm SVD-Type Latent Factor Models for Rating Prediction},
journal = {Algorithms},
year = {2026},
volume = {19},
number = {7},
pages = {600},
doi = {10.3390/a19070600}
}
EUVIMED