Randomized nonnegative matrix factorization

•A novel randomized hierarchical alternating least squares algorithm for NMF.•The randomized algorithm scales up to big data.•The algorithm outperforms previous compressed NMF algorithms in speed and accuracy.•Both synthetic and real-world data are used for evaluation.•A Python implementation is pro...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Pattern recognition letters Jg. 104; S. 1 - 7
Hauptverfasser: Erichson, N. Benjamin, Mendible, Ariana, Wihlborn, Sophie, Kutz, J. Nathan
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Amsterdam Elsevier B.V 01.03.2018
Elsevier Science Ltd
Schlagworte:
Algorithms
Big Data
Data management
Data mining
Data processing
Decomposition
Dimension reduction
Factorization
Matrix
NMF
Numerical analysis
Probability
Randomization
Randomized algorithm
Randomized algorithms
Randomized algorithm
Dimension reduction
NMF
ISSN:0167-8655, 1872-7344
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:•A novel randomized hierarchical alternating least squares algorithm for NMF.•The randomized algorithm scales up to big data.•The algorithm outperforms previous compressed NMF algorithms in speed and accuracy.•Both synthetic and real-world data are used for evaluation.•A Python implementation is provided on GitHub. Nonnegative matrix factorization (NMF) is a powerful tool for data mining. However, the emergence of ‘big data’ has severely challenged our ability to compute this fundamental decomposition using deterministic algorithms. This paper presents a randomized hierarchical alternating least squares (HALS) algorithm to compute the NMF. By deriving a smaller matrix from the nonnegative input data, a more efficient nonnegative decomposition can be computed. Our algorithm scales to big data applications while attaining a near-optimal factorization, i.e., the algorithm scales with the target rank of the data rather than the ambient dimension of measurement space. The proposed algorithm is evaluated using synthetic and real world data and shows substantial speedups compared to deterministic HALS.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0167-8655
1872-7344
DOI:10.1016/j.patrec.2018.01.007