Recovering the Number of Clusters From a Laplacian Matrix by Nuclear Norm Penalization
Statistical Analysis and Data Mining: An ASA Data Science Journal · 2025-08 · 18 (4)
DOI: 10.1002/sam.70031 ↗Kurz gefasst
Fachartikel in der Zeitschrift Statistical Analysis and Data Mining: An ASA Data Science Journal, online erschienen am 02.07.2025. Verfasst von Cinzia Di Nuzzo und 1 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 19 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Ministero dell'Università e della Ricerca.
Abstract (Originalsprache)
Spectral clustering is a powerful technique for data partitioning, but determining the optimal number of clusters remains challenging. This article introduces ALLE (ALgebraic Laplacian Estimator), an automatic method for estimating the number of clusters within the spectral clustering framework. By formulating the cluster recovery problem as a penalized minimization task, ALLE is able to systematically recover the number of clusters and the embedding space by assuming for the Laplacian matrix a low‐rank plus sparse decomposition. Specifically, ALLE recovers the low‐rank representation of the Laplacian matrix using nuclear norm plus ‐norm penalization. ALLE is computed via a proximal gradient algorithm alternating Singular Value Thresholding and Soft Thresholding, and it's very good performance is shown via a simulation study.
Förderung
- Ministero dell'Università e della Ricerca
Zitieren
APA
Nuzzo, C. D., & Farnè, M. (2025). Recovering the Number of Clusters From a Laplacian Matrix by Nuclear Norm Penalization. Statistical Analysis and Data Mining: An ASA Data Science Journal, 18(4). https://doi.org/10.1002/sam.70031
Vancouver
Nuzzo CD, Farnè M. Recovering the Number of Clusters From a Laplacian Matrix by Nuclear Norm Penalization. Statistical Analysis and Data Mining: An ASA Data Science Journal. 2025;18(4). doi:10.1002/sam.70031
BibTeX
@article{Nuzzo2025,
author = {Nuzzo, Cinzia Di and Farnè, Matteo},
title = {Recovering the Number of Clusters From a Laplacian Matrix by Nuclear Norm Penalization},
journal = {Statistical Analysis and Data Mining: An ASA Data Science Journal},
year = {2025},
volume = {18},
number = {4},
doi = {10.1002/sam.70031}
}
EUVIMED