Nicht-Negativ Matrixfaktorisierung (NMF) is a computational technique in the field of maschinellem Lernen and Datenanalyse. It involves decomposing a given non-negative matrix into two non-negative matrices, typically referred to as basis and coefficient matrices. The purpose of this factorization is to identify hidden patterns or structures within the data, making it particularly useful for tasks such as Themenmodellierung, der Bildverarbeitung, and kollaboratives Filtern.
Mathematisch betrachtet, gegeben eine nicht-negative Matrix V (with dimensions m x n), sucht NMF nach zwei nicht-negativen Matrizen W (basis matrix, of dimensions m x r) und H (coefficient matrix, of dimensions r x n), so dass:
V ≈ W * H
Hier ist r is the rank or number of components to be extracted, and the approximation seeks to minimize the difference between the original matrix and the product of the two factorized matrices. One of the key properties of NMF is that it allows for a parts-based representation of the data, as all components are constrained to be non-negative, leading to more interpretable results. This is in contrast to other matrix factorization techniques, such as Singular Value Zerlegung (SVD), which can yield negative values and thus may be less intuitive for certain applications.
NMF has applications across various domains, including image compression, document clustering, and Empfehlungssystemen, where it helps in extracting meaningful features from complex datasets. Its simplicity and effectiveness in revealing latent structures make it a popular choice among data scientists and researchers.