Non--negative matrix factorization using the \textit{R} package \textsf{nnmf}

2026-07-22

AI Summary

This study addresses the lack of systematic, real-data-based evaluation among existing Non-negative Matrix Factorization (NMF) implementations in R, which hinders informed package selection by users. To resolve this gap, we introduce nnmf, a new R package for NMF, and present the first comprehensive benchmark comparing nnmf against two widely used NMF packages within a unified experimental framework. The evaluation leverages real-world datasets and assesses performance across multiple dimensions—including computational efficiency, convergence behavior, reconstruction accuracy, memory consumption, and numerical stability. Our results clearly delineate the relative strengths and weaknesses of each implementation and demonstrate that nnmf achieves superior overall performance, thereby offering practitioners a reliable basis for selecting an appropriate NMF tool for real-world applications.

Abstract

Non--negative matrix factorization (NMF) has become an established dimensionality reduction technique for extracting latent structures from non--negative data and has found widespread applications in fields such as bioinformatics, text mining, image analysis, and recommender systems. As the popularity of NMF has increased, numerous \textit{R} packages implementing different optimization strategies and computational frameworks have been developed. Despite their widespread availability, comprehensive evaluations of these implementations under real--world data conditions remain limited. Consequently, researchers often lack objective guidance when selecting an appropriate package for practical applications. This study introduces a new \textit{R} package for NMF and offers asystematic performance comparison with two widely available \textit{R} packages for NMF analysis. Rather than relying on simulated datasets, the evaluation is conducted using real--world data to better reflect the complexity, heterogeneity, and noise characteristics encountered in practical analytical settings. The packages are assessed using a consistent experimental framework, with emphasis on computational efficiency, convergence behavior, reconstruction accuracy, memory utilization, and the stability of the resulting matrix factorization.

Problem

• Non-negative matrix factorization
• R package
• performance evaluation
• real-world data
• computational efficiency

Innovation

• non-negative matrix factorization
• R package
• real-world data evaluation
• performance benchmarking
• computational efficiency