Non-negative matrix factorization

Non-negative matrix factorization

(NMF or NNMF), also non-negative
matrix approximation[1][2] is a group
of algorithms in multivariate analysis
and linear algebra where a matrix V is
factorized into (usually) two matrices
W and H, with the property that all Illustration of approximate non-negative matrix factorization: the
three matrices have no negative matrix V is represented by the two smaller matrices W and H,
elements. This non-negativity makes which, when multiplied, approximately reconstruct V .
the resulting matrices easier to inspect.
Also, in applications such as
processing of audio spectrograms or muscular activity, non-negativity is inherent to the data being
considered. Since the problem is not exactly solvable in general, it is commonly approximated numerically.

NMF finds applications in such fields as astronomy,[3][4] computer vision, document clustering,[1] missing
data imputation,[5] chemometrics, audio signal processing, recommender systems,[6][7] and
bioinformatics.[8]

History
In chemometrics non-negative matrix factorization has a long history under the name "self modeling curve
resolution".[9] In this framework the vectors in the right matrix are continuous curves rather than discrete
vectors. Also early work on non-negative matrix factorizations was performed by a Finnish group of
researchers in the 1990s under the name positive matrix factorization.[10][11][12] It became more widely
known as non-negative matrix factorization after Lee and Seung investigated the properties of the
algorithm and published some simple and useful algorithms for two types of factorizations.[13][14]

Background
Let matrix V be the product of the matrices W and H,

Matrix multiplication can be implemented as computing the column vectors of V as linear combinations of
the column vectors in W using coefficients supplied by columns of H. That is, each column of V can be
computed as follows:

where v i is the i-th column vector of the product matrix V and h i is the i-th column vector of the matrix
H.
When multiplying matrices, the dimensions of the factor matrices may be significantly lower than those of
the product matrix and it is this property that forms the basis of NMF. NMF generates factors with
significantly reduced dimensions compared to the original matrix. For example, if V is an m × n matrix,
W is an m × p matrix, and H is a p × n matrix then p can be significantly less than both m and n.
Here is an example based on a text-mining application:

Let the input matrix (the matrix to be factored) be V with 10000 rows and 500 columns where
words are in rows and documents are in columns. That is, we have 500 documents indexed
by 10000 words. It follows that a column vector v in V represents a document.
Assume we ask the algorithm to find 10 features in order to generate a features matrix W
with 10000 rows and 10 columns and a coefficients matrix H with 10 rows and 500 columns.
The product of W and H is a matrix with 10000 rows and 500 columns, the same shape as
the input matrix V and, if the factorization worked, it is a reasonable approximation to the
input matrix V.
From the treatment of matrix multiplication above it follows that each column in the product
matrix WH is a linear combination of the 10 column vectors in the features matrix W with
coefficients supplied by the coefficients matrix H.

This last point is the basis of NMF because we can consider each original document in our example as
being built from a small set of hidden features. NMF generates these features.

It is useful to think of each feature (column vector) in the features matrix W as a document archetype
comprising a set of words where each word's cell value defines the word's rank in the feature: The higher a
word's cell value the higher the word's rank in the feature. A column in the coefficients matrix H represents
an original document with a cell value defining the document's rank for a feature. We can now reconstruct a
document (column vector) from our input matrix by a linear combination of our features (column vectors in
W) where each feature is weighted by the feature's cell value from the document's column in H.

Clustering property
NMF has an inherent clustering property,[15] i.e., it automatically clusters the columns of input data
.

More specifically, the approximation of by is achieved by finding and that minimize

the error function (using the Frobenius norm)

subject to ,

If we furthermore impose an orthogonality constraint on , i.e. , then the above minimization is

mathematically equivalent to the minimization of K-means clustering.[15]

Furthermore, the computed gives the cluster membership, i.e., if for all i ≠ k, this suggests
that the input data belongs to -th cluster. The computed gives the cluster centroids, i.e., the -th
column gives the cluster centroid of -th cluster. This centroid's representation can be significantly
enhanced by convex NMF.

When the orthogonality constraint is not explicitly imposed, the orthogonality holds to a large
extent, and the clustering property holds too. Clustering is the main objective of most data mining
applications of NMF.
When the error function to be used is Kullback–Leibler divergence, NMF is identical to the probabilistic
latent semantic analysis (PLSA), a popular document clustering method.[16]

Types

Approximate non-negative matrix factorization

Usually the number of columns of W and the number of rows of H in NMF are selected so the product
WH will become an approximation to V. The full decomposition of V then amounts to the two non-
negative matrices W and H as well as a residual U, such that: V = WH + U. The elements of the
residual matrix can either be negative or positive.

When W and H are smaller than V they become easier to store and manipulate. Another reason for
factorizing V into smaller matrices W and H, is that if one is able to approximately represent the elements
of V by significantly less data, then one has to infer some latent structure in the data.

Convex non-negative matrix factorization

，
In standard NMF, matrix factor W ∈ R+m × k i.e., W can be anything in that space. Convex NMF[17]
restricts the columns of W to convex combinations of the input data vectors . This greatly
improves the quality of data representation of W. Furthermore, the resulting matrix factor H becomes more
sparse and orthogonal.

Nonnegative rank factorization

In case the nonnegative rank of V is equal to its actual rank, V = WH is called a nonnegative rank
factorization (NRF).[18][19][20] The problem of finding the NRF of V, if it exists, is known to be NP-
hard.[21]

Different cost functions and regularizations

There are different types of non-negative matrix factorizations. The different types arise from using
different cost functions for measuring the divergence between V and WH and possibly by regularization
of the W and/or H matrices.[1]

Two simple divergence functions studied by Lee and Seung are the squared error (or Frobenius norm) and
an extension of the Kullback–Leibler divergence to positive matrices (the original Kullback–Leibler
divergence is defined on probability distributions). Each divergence leads to a different NMF algorithm,
usually minimizing the divergence using iterative update rules.

The factorization problem in the squared error version of NMF may be stated as: Given a matrix find
nonnegative matrices W and H that minimize the function

Another type of NMF for images is based on the total variation norm.[22]
When L1 regularization (akin to Lasso) is added to NMF with the mean squared error cost function, the
resulting problem may be called non-negative sparse coding due to the similarity to the sparse coding
problem,[23][24] although it may also still be referred to as NMF.[25]

Online NMF

Many standard NMF algorithms analyze all the data together; i.e., the whole matrix is available from the
start. This may be unsatisfactory in applications where there are too many data to fit into memory or where
the data are provided in streaming fashion. One such use is for collaborative filtering in recommendation
systems, where there may be many users and many items to recommend, and it would be inefficient to
recalculate everything when one user or one item is added to the system. The cost function for optimization
in these cases may or may not be the same as for standard NMF, but the algorithms need to be rather
different.[26][27]

Convolutional NMF

If the columns of V represent data sampled over spatial or temporal dimensions, e.g. time signals, images,
or video, features that are equivariant w.r.t. shifts along these dimensions can be learned by Convolutional
NMF. In this case, W is sparse with columns having local non-zero weight windows that are shared across
shifts along the spatio-temporal dimensions of V, representing convolution kernels. By spatio-temporal
pooling of H and repeatedly using the resulting representation as input to convolutional NMF, deep feature
hierarchies can be learned.[28]

Algorithms
There are several ways in which the W and H may be found: Lee and Seung's multiplicative update
rule[14] has been a popular method due to the simplicity of implementation. This algorithm is:

initialize: W and H non negative.

Then update the values in W and H by computing the following, with as an index of the
iteration.

and

Until W and H are stable.

Note that the updates are done on an element by element basis not matrix multiplication.

We note that the multiplicative factors for W and H, i.e. the and terms, are matrices of
ones when .

More recently other algorithms have been developed. Some approaches are based on alternating non-
negative least squares: in each step of such an algorithm, first H is fixed and W found by a non-negative
least squares solver, then W is fixed and H is found analogously. The procedures used to solve for W and
H may be the same[29] or different, as some NMF variants regularize one of W and H.[23] Specific
approaches include the projected gradient descent methods,[29][30] the active set method,[6][31] the optimal
gradient method,[32] and the block principal pivoting method[33] among several others.[34]

Current algorithms are sub-optimal in that they only guarantee finding a local minimum, rather than a global
minimum of the cost function. A provably optimal algorithm is unlikely in the near future as the problem
has been shown to generalize the k-means clustering problem which is known to be NP-complete.[35]
However, as in many other data mining applications, a local minimum may still prove to be useful.

Sequential NMF

Fractional residual variance (FRV) plots for PCA and sequential NMF;[4] for PCA, the
theoretical values are the contribution from the residual eigenvalues. In comparison,
the FRV curves for PCA reaches a flat plateau where no signal are captured
effectively; while the NMF FRV curves are declining continuously, indicating a better
ability to capture signal. The FRV curves for NMF also converges to higher levels
than PCA, indicating the less-overfitting property of NMF.

The sequential construction of NMF components (W and H) was firstly used to relate NMF with Principal
Component Analysis (PCA) in astronomy.[36] The contribution from the PCA components are ranked by
the magnitude of their corresponding eigenvalues; for NMF, its components can be ranked empirically
when they are constructed one by one (sequentially), i.e., learn the -th component with the first
components constructed.

The contribution of the sequential NMF components can be compared with the Karhunen–Loève theorem,
an application of PCA, using the plot of eigenvalues. A typical choice of the number of components with
PCA is based on the "elbow" point, then the existence of the flat plateau is indicating that PCA is not
capturing the data efficiently, and at last there exists a sudden drop reflecting the capture of random noise
and falls into the regime of overfitting.[37][38] For sequential NMF, the plot of eigenvalues is approximated
by the plot of the fractional residual variance curves, where the curves decreases continuously, and
converge to a higher level than PCA,[4] which is the indication of less over-fitting of sequential NMF.

Exact NMF

Exact solutions for the variants of NMF can be expected (in polynomial time) when additional constraints
hold for matrix V. A polynomial time algorithm for solving nonnegative rank factorization if V contains a
monomial sub matrix of rank equal to its rank was given by Campbell and Poole in 1981.[39] Kalofolias
and Gallopoulos (2012)[40] solved the symmetric counterpart of this problem, where V is symmetric and
contains a diagonal principal sub matrix of rank r. Their algorithm runs in O(rm2) time in the dense case.
Arora, Ge, Halpern, Mimno, Moitra, Sontag, Wu, & Zhu (2013) give a polynomial time algorithm for exact
NMF that works for the case where one of the factors W satisfies a separability condition.[41]

Relation to other techniques

In Learning the parts of objects by non-negative matrix factorization Lee and Seung[42] proposed NMF
mainly for parts-based decomposition of images. It compares NMF to vector quantization and principal
component analysis, and shows that although the three techniques may be written as factorizations, they
implement different constraints and therefore produce different results.

It was later shown that some types of NMF are an instance of a

more general probabilistic model called "multinomial PCA".[43]
When NMF is obtained by minimizing the Kullback–Leibler
divergence, it is in fact equivalent to another instance of
multinomial PCA, probabilistic latent semantic analysis,[44] trained
by maximum likelihood estimation. That method is commonly used
for analyzing and clustering textual data and is also related to the
latent class model.

NMF with the least-squares objective is equivalent to a relaxed

form of K-means clustering: the matrix factor W contains cluster
centroids and H contains cluster membership indicators.[15][45]
This provides a theoretical foundation for using NMF for data
clustering. However, k-means does not enforce non-negativity on NMF as a probabilistic graphical
its centroids, so the closest analogy is in fact with "semi-NMF".[17] model: visible units (V ) are
connected to hidden units (H)
NMF can be seen as a two-layer directed graphical model with one through weights W, so that V is
layer of observed random variables and one layer of hidden random generated from a probability
variables.[46] distribution with mean

NMF extends beyond matrices to tensors of arbitrary .[13]: 5 

order.[47][48][49] This extension may be viewed as a non-negative

counterpart to, e.g., the PARAFAC model.

Other extensions of NMF include joint factorization of several data matrices and tensors where some
factors are shared. Such models are useful for sensor fusion and relational learning.[50]

NMF is an instance of nonnegative quadratic programming (NQP), just like the support vector machine
(SVM). However, SVM and NMF are related at a more intimate level than that of NQP, which allows
direct application of the solution algorithms developed for either of the two methods to problems in both
domains.[51]

Uniqueness
The factorization is not unique: A matrix and its inverse can be used to transform the two factorization
matrices by, e.g.,[52]
If the two new matrices and are non-negative they form another parametrization
of the factorization.

The non-negativity of and applies at least if B is a non-negative monomial matrix. In this simple
case it will just correspond to a scaling and a permutation.

More control over the non-uniqueness of NMF is obtained with sparsity constraints.[53]

Applications

Astronomy

In astronomy, NMF is a promising method for dimension reduction in the sense that astrophysical signals
are non-negative. NMF has been applied to the spectroscopic observations[54][3] and the direct imaging
observations[4] as a method to study the common properties of astronomical objects and post-process the
astronomical observations. The advances in the spectroscopic observations by Blanton & Roweis (2007)[3]
takes into account of the uncertainties of astronomical observations, which is later improved by Zhu
(2016)[36] where missing data are also considered and parallel computing is enabled. Their method is then
adopted by Ren et al. (2018)[4] to the direct imaging field as one of the methods of detecting exoplanets,
especially for the direct imaging of circumstellar disks.

Ren et al. (2018)[4] are able to prove the stability of NMF components when they are constructed
sequentially (i.e., one by one), which enables the linearity of the NMF modeling process; the linearity
property is used to separate the stellar light and the light scattered from the exoplanets and circumstellar
disks.

In direct imaging, to reveal the faint exoplanets and circumstellar disks from bright the surrounding stellar
lights, which has a typical contrast from 10⁵ to 10¹⁰, various statistical methods have been
adopted,[55][56][37] however the light from the exoplanets or circumstellar disks are usually over-fitted,
where forward modeling have to be adopted to recover the true flux.[57][38] Forward modeling is currently
optimized for point sources,[38] however not for extended sources, especially for irregularly shaped
structures such as circumstellar disks. In this situation, NMF has been an excellent method, being less over-
fitting in the sense of the non-negativity and sparsity of the NMF modeling coefficients, therefore forward
modeling can be performed with a few scaling factors,[4] rather than a computationally intensive data re-
reduction on generated models.

Data imputation

To impute missing data in statistics, NMF can take missing data while minimizing its cost function, rather
than treating these missing data as zeros.[5] This makes it a mathematically proven method for data
imputation in statistics.[5] By first proving that the missing data are ignored in the cost function, then
proving that the impact from missing data can be as small as a second order effect, Ren et al. (2020)[5]
studied and applied such an approach for the field of astronomy. Their work focuses on two-dimensional
matrices, specifically, it includes mathematical derivation, simulated data imputation, and application to on-
sky data.

The data imputation procedure with NMF can be composed of two steps. First, when the NMF
components are known, Ren et al. (2020) proved that impact from missing data during data imputation
("target modeling" in their study) is a second order effect. Second, when the NMF components are
unknown, the authors proved that the impact from missing data during component construction is a first-to-
second order effect.

Depending on the way that the NMF components are obtained, the former step above can be either
independent or dependent from the latter. In addition, the imputation quality can be increased when the
more NMF components are used, see Figure 4 of Ren et al. (2020) for their illustration.[5]

Text mining

NMF can be used for text mining applications. In this process, a document-term matrix is constructed with
the weights of various terms (typically weighted word frequency information) from a set of documents.
This matrix is factored into a term-feature and a feature-document matrix. The features are derived from the
contents of the documents, and the feature-document matrix describes data clusters of related documents.

One specific application used hierarchical NMF on a small subset of scientific abstracts from PubMed.[58]
Another research group clustered parts of the Enron email dataset[59] with 65,033 messages and 91,133
terms into 50 clusters.[60] NMF has also been applied to citations data, with one example clustering English
Wikipedia articles and scientific journals based on the outbound scientific citations in English
Wikipedia.[61]

Arora, Ge, Halpern, Mimno, Moitra, Sontag, Wu, & Zhu (2013) have given polynomial-time algorithms to
learn topic models using NMF. The algorithm assumes that the topic matrix satisfies a separability condition
that is often found to hold in these settings.[41]

Hassani, Iranmanesh and Mansouri (2019) proposed a feature agglomeration method for term-document
matrices which operates using NMF. The algorithm reduces the term-document matrix into a smaller matrix
more suitable for text clustering.[62]

Spectral data analysis

NMF is also used to analyze spectral data; one such use is in the classification of space objects and
debris.[63]

Scalable Internet distance prediction

NMF is applied in scalable Internet distance (round-trip time) prediction. For a network with hosts, with
the help of NMF, the distances of all the end-to-end links can be predicted after conducting only
measurements. This kind of method was firstly introduced in Internet Distance Estimation Service
(IDES).[64] Afterwards, as a fully decentralized approach, Phoenix network coordinate system[65] is
proposed. It achieves better overall prediction accuracy by introducing the concept of weight.

Non-stationary speech denoising

Speech denoising has been a long lasting problem in audio signal processing. There are many algorithms
for denoising if the noise is stationary. For example, the Wiener filter is suitable for additive Gaussian noise.
However, if the noise is non-stationary, the classical denoising algorithms usually have poor performance
because the statistical information of the non-stationary noise is difficult to estimate. Schmidt et al.[66] use
NMF to do speech denoising under non-stationary noise, which is completely different from classical
statistical approaches. The key idea is that clean speech signal can be sparsely represented by a speech
dictionary, but non-stationary noise cannot. Similarly, non-stationary noise can also be sparsely represented
by a noise dictionary, but speech cannot.

The algorithm for NMF denoising goes as follows. Two dictionaries, one for speech and one for noise,
need to be trained offline. Once a noisy speech is given, we first calculate the magnitude of the Short-Time-
Fourier-Transform. Second, separate it into two parts via NMF, one can be sparsely represented by the
speech dictionary, and the other part can be sparsely represented by the noise dictionary. Third, the part that
is represented by the speech dictionary will be the estimated clean speech.

Population genetics

Sparse NMF is used in Population genetics for estimating individual admixture coefficients, detecting
genetic clusters of individuals in a population sample or evaluating genetic admixture in sampled genomes.
In human genetic clustering, NMF algorithms provide estimates similar to those of the computer program
STRUCTURE, but the algorithms are more efficient computationally and allow analysis of large
population genomic data sets.[67]

Bioinformatics

NMF has been successfully applied in bioinformatics for clustering gene expression and DNA methylation
data and finding the genes most representative of the clusters.[24][68][69][70] In the analysis of cancer
mutations it has been used to identify common patterns of mutations that occur in many cancers and that
probably have distinct causes.[71] NMF techniques can identify sources of variation such as cell types,
disease subtypes, population stratification, tissue composition, and tumor clonality.[72]

A particular variant of NMF, namely Non-Negative Matrix Tri-Factorization (NMTF),[73] has been use for
drug repurposing tasks in order to predict novel protein targets and therapeutic indications for approved
drugs[74] and to infer pair of synergic anticancer drugs.[75]

Nuclear imaging

NMF, also referred in this field as factor analysis, has been used since the 1980s[76] to analyze sequences of
images in SPECT and PET dynamic medical imaging. Non-uniqueness of NMF was addressed using
sparsity constraints.[77] [78] [79]

Current research
Current research (since 2010) in nonnegative matrix factorization includes, but is not limited to,

1. Algorithmic: searching for global minima of the factors and factor initialization.[80]
2. Scalability: how to factorize million-by-billion matrices, which are commonplace in Web-
scale data mining, e.g., see Distributed Nonnegative Matrix Factorization (DNMF),[81]
Scalable Nonnegative Matrix Factorization (ScalableNMF),[82] Distributed Stochastic
Singular Value Decomposition.[83]
3. Online: how to update the factorization when new data comes in without recomputing from
scratch, e.g., see online CNSC[84]
4. Collective (joint) factorization: factorizing multiple interrelated matrices for multiple-view
learning, e.g. multi-view clustering, see CoNMF[85] and MultiNMF[86]
5. Cohen and Rothblum 1993 problem: whether a rational matrix always has an NMF of
minimal inner dimension whose factors are also rational. Recently, this problem has been
answered negatively.[87]

See also
Multilinear algebra
Multilinear subspace learning
Tensor
Tensor decomposition
Tensor software

Sources and external links

Notes
1. Suvrit Sra; Inderjit S. Dhillon (2006). Generalized Nonnegative Matrix Approximations with
Bregman Divergences (https://papers.nips.cc/paper/2757-generalized-nonnegative-matrix-a
pproximations-with-bregman-divergences.pdf) (PDF). Advances in Neural Information
Processing Systems 18. Advances in Neural Information Processing Systems. ISBN 978-0-
262-23253-1. Wikidata Q77685465.
2. Tandon, Rashish; Sra, Suvrit (September 13, 2010). Sparse nonnegative matrix
approximation: new formulations and algorithms (https://is.tuebingen.mpg.de/fileadmin/user_
upload/files/publications/MPIK-TR-193_%5B0%5D.pdf) (PDF) (Report). Max Planck Institute
for Biological Cybernetics. Technical Report No. 193.
3. Blanton, Michael R.; Roweis, Sam (2007). "K-corrections and filter transformations in the
ultraviolet, optical, and near infrared". The Astronomical Journal. 133 (2): 734–754.
arXiv:astro-ph/0606170 (https://arxiv.org/abs/astro-ph/0606170).
Bibcode:2007AJ....133..734B (https://ui.adsabs.harvard.edu/abs/2007AJ....133..734B).
doi:10.1086/510127 (https://doi.org/10.1086%2F510127). S2CID 18561804 (https://api.sem
anticscholar.org/CorpusID:18561804).
4. Ren, Bin; Pueyo, Laurent; Zhu, Guangtun B.; Duchêne, Gaspard (2018). "Non-negative
Matrix Factorization: Robust Extraction of Extended Structures". The Astrophysical Journal.
852 (2): 104. arXiv:1712.10317 (https://arxiv.org/abs/1712.10317).
Bibcode:2018ApJ...852..104R (https://ui.adsabs.harvard.edu/abs/2018ApJ...852..104R).
doi:10.3847/1538-4357/aaa1f2 (https://doi.org/10.3847%2F1538-4357%2Faaa1f2).
S2CID 3966513 (https://api.semanticscholar.org/CorpusID:3966513).
5. Ren, Bin; Pueyo, Laurent; Chen, Christine; Choquet, Elodie; Debes, John H; Duechene,
Gaspard; Menard, Francois; Perrin, Marshall D. (2020). "Using Data Imputation for Signal
Separation in High Contrast Imaging". The Astrophysical Journal. 892 (2): 74.
arXiv:2001.00563 (https://arxiv.org/abs/2001.00563). Bibcode:2020ApJ...892...74R (https://u
i.adsabs.harvard.edu/abs/2020ApJ...892...74R). doi:10.3847/1538-4357/ab7024 (https://doi.
org/10.3847%2F1538-4357%2Fab7024). S2CID 209531731 (https://api.semanticscholar.or
g/CorpusID:209531731).
6. Rainer Gemulla; Erik Nijkamp; Peter J. Haas; Yannis Sismanis (2011). Large-scale matrix
factorization with distributed stochastic gradient descent. Proc. ACM SIGKDD Int'l Conf. on
Knowledge discovery and data mining. pp. 69–77.
 7. Yang Bao; et al. (2014). TopicMF: Simultaneously Exploiting Ratings and Reviews for
Recommendation (http://www.aaai.org/ocs/index.php/AAAI/AAAI14/paper/view/8273). AAAI.
8. Ben Murrell; et al. (2011). "Non-Negative Matrix Factorization for Learning Alignment-
Specific Models of Protein Evolution" (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC32452
33). PLOS ONE. 6 (12): e28898. Bibcode:2011PLoSO...628898M (https://ui.adsabs.harvard.
edu/abs/2011PLoSO...628898M). doi:10.1371/journal.pone.0028898 (https://doi.org/10.137
1%2Fjournal.pone.0028898). PMC 3245233 (https://www.ncbi.nlm.nih.gov/pmc/articles/PM
C3245233). PMID 22216138 (https://pubmed.ncbi.nlm.nih.gov/22216138).
9. William H. Lawton; Edward A. Sylvestre (1971). "Self modeling curve resolution".
Technometrics. 13 (3): 617–633. doi:10.2307/1267173 (https://doi.org/10.2307%2F126717
3). JSTOR 1267173 (https://www.jstor.org/stable/1267173).
10. Pentti Paatero; Unto Tapper; Pasi Aalto; Markku Kulmala (1991). "Matrix factorization
methods for analysing diffusion battery data". Journal of Aerosol Science. 22: S273–S276.
doi:10.1016/S0021-8502(05)80089-8 (https://doi.org/10.1016%2FS0021-8502%2805%2980
089-8). ISSN 0021-8502 (https://www.worldcat.org/issn/0021-8502). Wikidata Q58065673.
11. Pentti Paatero; Unto Tapper (June 1994). "Positive matrix factorization: A non-negative factor
model with optimal utilization of error estimates of data values" (https://onlinelibrary.wiley.co
m/doi/abs/10.1002/env.3170050203). Environmetrics. 5 (2): 111–126.
doi:10.1002/ENV.3170050203 (https://doi.org/10.1002%2FENV.3170050203). ISSN 1180-
4009 (https://www.worldcat.org/issn/1180-4009). Wikidata Q29308406.
12. Pia Anttila; Pentti Paatero; Unto Tapper; Olli Järvinen (1995). "Source identification of bulk
wet deposition in Finland by positive matrix factorization". Atmospheric Environment. 29
(14): 1705–1718. Bibcode:1995AtmEn..29.1705A (https://ui.adsabs.harvard.edu/abs/1995At
mEn..29.1705A). doi:10.1016/1352-2310(94)00367-T (https://doi.org/10.1016%2F1352-231
0%2894%2900367-T).
13. Daniel D. Lee & H. Sebastian Seung (1999). "Learning the parts of objects by non-negative
matrix factorization". Nature. 401 (6755): 788–791. Bibcode:1999Natur.401..788L (https://ui.a
dsabs.harvard.edu/abs/1999Natur.401..788L). doi:10.1038/44565 (https://doi.org/10.1038%2
F44565). PMID 10548103 (https://pubmed.ncbi.nlm.nih.gov/10548103). S2CID 4428232 (htt
ps://api.semanticscholar.org/CorpusID:4428232).
14. Daniel D. Lee & H. Sebastian Seung (2001). Algorithms for Non-negative Matrix
Factorization (http://papers.nips.cc/paper/1861-algorithms-for-non-negative-matrix-factorizati
on.pdf) (PDF). Advances in Neural Information Processing Systems 13: Proceedings of the
2000 Conference. MIT Press. pp. 556–562.
15. C. Ding, X. He, H.D. Simon (2005). "On the Equivalence of Nonnegative Matrix Factorization
and Spectral Clustering" (http://ranger.uta.edu/~chqding/papers/NMF-SDM2005.pdf). Proc.
SIAM Int'l Conf. Data Mining, pp. 606-610. May 2005
16. Ding C, Li Y, Peng W (2008). "On the equivalence between non-negative matrix factorization
and probabilistic latent semantic indexing" (https://web.archive.org/web/20160304070027/ht
tp://users.cis.fiu.edu/~taoli/pub/NMFpLSIequiv.pdf) (PDF). Computational Statistics & Data
Analysis. 52 (8): 3913–3927. doi:10.1016/j.csda.2008.01.011 (https://doi.org/10.1016%2Fj.c
sda.2008.01.011). Archived from the original (http://users.cis.fiu.edu/~taoli/pub/NMFpLSIequi
v.pdf) (PDF) on 2016-03-04.
17. C Ding, T Li, MI Jordan, Convex and semi-nonnegative matrix factorizations, IEEE
Transactions on Pattern Analysis and Machine Intelligence, 32, 45-55, 2010
18. Berman, A.; R.J. Plemmons (1974). "Inverses of nonnegative matrices". Linear and
Multilinear Algebra. 2 (2): 161–172. doi:10.1080/03081087408817055 (https://doi.org/10.108
0%2F03081087408817055).
19. A. Berman; R.J. Plemmons (1994). Nonnegative matrices in the Mathematical Sciences.
Philadelphia: SIAM.
20. Thomas, L.B. (1974). "Problem 73-14, Rank factorization of nonnegative matrices". SIAM
Rev. 16 (3): 393–394. doi:10.1137/1016064 (https://doi.org/10.1137%2F1016064).
21. Vavasis, S.A. (2009). "On the complexity of nonnegative matrix factorization". SIAM J. Optim.
20 (3): 1364–1377. arXiv:0708.4149 (https://arxiv.org/abs/0708.4149).
doi:10.1137/070709967 (https://doi.org/10.1137%2F070709967). S2CID 7150400 (https://ap
i.semanticscholar.org/CorpusID:7150400).
22. Zhang, T.; Fang, B.; Liu, W.; Tang, Y. Y.; He, G.; Wen, J. (2008). "Total variation norm-based
nonnegative matrix factorization for identifying discriminant representation of image
patterns". Neurocomputing. 71 (10–12): 1824–1831. doi:10.1016/j.neucom.2008.01.022 (htt
ps://doi.org/10.1016%2Fj.neucom.2008.01.022).
23. Hoyer, Patrik O. (2002). Non-negative sparse coding. Proc. IEEE Workshop on Neural
Networks for Signal Processing. arXiv:cs/0202009 (https://arxiv.org/abs/cs/0202009).
24. Leo Taslaman & Björn Nilsson (2012). "A framework for regularized non-negative matrix
factorization, with application to the analysis of gene expression data" (https://www.ncbi.nlm.
nih.gov/pmc/articles/PMC3487913). PLOS One. 7 (11): e46331.
Bibcode:2012PLoSO...746331T (https://ui.adsabs.harvard.edu/abs/2012PLoSO...746331T).
doi:10.1371/journal.pone.0046331 (https://doi.org/10.1371%2Fjournal.pone.0046331).
PMC 3487913 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3487913). PMID 23133590
(https://pubmed.ncbi.nlm.nih.gov/23133590).
25. Hsieh, C. J.; Dhillon, I. S. (2011). Fast coordinate descent methods with variable selection
for non-negative matrix factorization (http://www.cs.utexas.edu/~cjhsieh/nmf_kdd11.pdf)
(PDF). Proceedings of the 17th ACM SIGKDD international conference on Knowledge
discovery and data mining - KDD '11. p. 1064. doi:10.1145/2020408.2020577 (https://doi.or
g/10.1145%2F2020408.2020577). ISBN 9781450308137.
26. Fung, Yik-Hing; Li, Chun-Hung; Cheung, William K. (2 November 2007). Online Discussion
Participation Prediction Using Non-negative Matrix Factorization (http://dl.acm.org/citation.cf
m?id=1339264.1339709). Wi-Iatw '07. IEEE Computer Society. pp. 284–287.
ISBN 9780769530284 – via dl.acm.org.
27. Naiyang Guan; Dacheng Tao; Zhigang Luo & Bo Yuan (July 2012). "Online Nonnegative
Matrix Factorization With Robust Stochastic Approximation". IEEE Transactions on Neural
Networks and Learning Systems. 23 (7): 1087–1099. doi:10.1109/TNNLS.2012.2197827 (htt
ps://doi.org/10.1109%2FTNNLS.2012.2197827). PMID 24807135 (https://pubmed.ncbi.nlm.
nih.gov/24807135). S2CID 8755408 (https://api.semanticscholar.org/CorpusID:8755408).
28. Behnke, S. (2003). "Discovering hierarchical speech features using convolutional non-
negative matrix factorization" (https://ieeexplore.ieee.org/document/1224004). Proceedings
of the International Joint Conference on Neural Networks, 2003. Portland, Oregon USA:
IEEE. 4: 2758–2763. doi:10.1109/IJCNN.2003.1224004 (https://doi.org/10.1109%2FIJCNN.
2003.1224004). ISBN 978-0-7803-7898-8. S2CID 3109867 (https://api.semanticscholar.org/
CorpusID:3109867).
29. Lin, Chih-Jen (2007). "Projected Gradient Methods for Nonnegative Matrix Factorization" (htt
p://www.csie.ntu.edu.tw/~cjlin/papers/pgradnmf.pdf) (PDF). Neural Computation. 19 (10):
2756–2779. CiteSeerX 10.1.1.308.9135 (https://citeseerx.ist.psu.edu/viewdoc/summary?doi
=10.1.1.308.9135). doi:10.1162/neco.2007.19.10.2756 (https://doi.org/10.1162%2Fneco.200
7.19.10.2756). PMID 17716011 (https://pubmed.ncbi.nlm.nih.gov/17716011).
S2CID 2295736 (https://api.semanticscholar.org/CorpusID:2295736).
30. Lin, Chih-Jen (2007). "On the Convergence of Multiplicative Update Algorithms for
Nonnegative Matrix Factorization". IEEE Transactions on Neural Networks. 18 (6): 1589–
1596. CiteSeerX 10.1.1.407.318 (https://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.
407.318). doi:10.1109/TNN.2007.895831 (https://doi.org/10.1109%2FTNN.2007.895831).
S2CID 2183630 (https://api.semanticscholar.org/CorpusID:2183630).
31. Hyunsoo Kim & Haesun Park (2008). "Nonnegative Matrix Factorization Based on
Alternating Nonnegativity Constrained Least Squares and Active Set Method" (http://www.c
c.gatech.edu/~hpark/papers/simax-nmf.pdf) (PDF). SIAM Journal on Matrix Analysis and
Applications. 30 (2): 713–730. CiteSeerX 10.1.1.70.3485 (https://citeseerx.ist.psu.edu/viewd
oc/summary?doi=10.1.1.70.3485). doi:10.1137/07069239x (https://doi.org/10.1137%2F0706
9239x).
32. Naiyang Guan; Dacheng Tao; Zhigang Luo; Bo Yuan (June 2012). "NeNMF: An Optimal
Gradient Method for Nonnegative Matrix Factorization". IEEE Transactions on Signal
Processing. 60 (6): 2882–2898. Bibcode:2012ITSP...60.2882G (https://ui.adsabs.harvard.ed
u/abs/2012ITSP...60.2882G). doi:10.1109/TSP.2012.2190406 (https://doi.org/10.1109%2FT
SP.2012.2190406). S2CID 8143231 (https://api.semanticscholar.org/CorpusID:8143231).
33. Jingu Kim & Haesun Park (2011). "Fast Nonnegative Matrix Factorization: An Active-set-like
Method and Comparisons". SIAM Journal on Scientific Computing. 58 (6): 3261–3281.
CiteSeerX 10.1.1.419.798 (https://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.419.7
98). doi:10.1137/110821172 (https://doi.org/10.1137%2F110821172).
34. Jingu Kim; Yunlong He & Haesun Park (2013). "Algorithms for nonnegative matrix and
tensor factorizations: A unified view based on block coordinate descent framework" (https://s
mallk.github.io/papers/nmf_review_jgo.pdf) (PDF). Journal of Global Optimization. 33 (2):
285–319. doi:10.1007/s10898-013-0035-4 (https://doi.org/10.1007%2Fs10898-013-0035-4).
S2CID 11197117 (https://api.semanticscholar.org/CorpusID:11197117).
35. Ding, C.; He, X. & Simon, H.D. (2005). "On the equivalence of nonnegative matrix
factorization and spectral clustering". Proc. SIAM Data Mining Conf. Vol. 4. pp. 606–610.
doi:10.1137/1.9781611972757.70 (https://doi.org/10.1137%2F1.9781611972757.70).
ISBN 978-0-89871-593-4.
36. Zhu, Guangtun B. (2016-12-19). "Nonnegative Matrix Factorization (NMF) with
Heteroscedastic Uncertainties and Missing data". arXiv:1612.06037 (https://arxiv.org/abs/16
12.06037) [astro-ph.IM (https://arxiv.org/archive/astro-ph.IM)].
37. Soummer, Rémi; Pueyo, Laurent; Larkin, James (2012). "Detection and Characterization of
Exoplanets and Disks Using Projections on Karhunen-Loève Eigenimages". The
Astrophysical Journal Letters. 755 (2): L28. arXiv:1207.4197 (https://arxiv.org/abs/1207.419
7). Bibcode:2012ApJ...755L..28S (https://ui.adsabs.harvard.edu/abs/2012ApJ...755L..28S).
doi:10.1088/2041-8205/755/2/L28 (https://doi.org/10.1088%2F2041-8205%2F755%2F2%2F
L28). S2CID 51088743 (https://api.semanticscholar.org/CorpusID:51088743).
38. Pueyo, Laurent (2016). "Detection and Characterization of Exoplanets using Projections on
Karhunen Loeve Eigenimages: Forward Modeling". The Astrophysical Journal. 824 (2): 117.
arXiv:1604.06097 (https://arxiv.org/abs/1604.06097). Bibcode:2016ApJ...824..117P (https://u
i.adsabs.harvard.edu/abs/2016ApJ...824..117P). doi:10.3847/0004-637X/824/2/117 (https://d
oi.org/10.3847%2F0004-637X%2F824%2F2%2F117). S2CID 118349503 (https://api.seman
ticscholar.org/CorpusID:118349503).
39. Campbell, S.L.; G.D. Poole (1981). "Computing nonnegative rank factorizations" (https://doi.
org/10.1016%2F0024-3795%2881%2990272-x). Linear Algebra Appl. 35: 175–182.
doi:10.1016/0024-3795(81)90272-x (https://doi.org/10.1016%2F0024-3795%2881%299027
2-x).
40. Kalofolias, V.; Gallopoulos, E. (2012). "Computing symmetric nonnegative rank
factorizations" (https://infoscience.epfl.ch/record/198764/files/main.pdf) (PDF). Linear
Algebra Appl. 436 (2): 421–435. doi:10.1016/j.laa.2011.03.016 (https://doi.org/10.1016%2Fj.l
aa.2011.03.016).
41. Arora, Sanjeev; Ge, Rong; Halpern, Yoni; Mimno, David; Moitra, Ankur; Sontag, David; Wu,
Yichen; Zhu, Michael (2013). A practical algorithm for topic modeling with provable
guarantees (http://jmlr.csail.mit.edu/proceedings/papers/v28/arora13.html). Proceedings of
the 30th International Conference on Machine Learning. arXiv:1212.4777 (https://arxiv.org/a
bs/1212.4777). Bibcode:2012arXiv1212.4777A (https://ui.adsabs.harvard.edu/abs/2012arXi
v1212.4777A).
42. Lee, Daniel D.; Sebastian, Seung, H. (1999). "Learning the parts of objects by non-negative
matrix factorization" (http://www.columbia.edu/~jwp2128/Teaching/E4903/papers/nmf_natur
e.pdf) (PDF). Nature. 401 (6755): 788–791. Bibcode:1999Natur.401..788L (https://ui.adsabs.
harvard.edu/abs/1999Natur.401..788L). doi:10.1038/44565 (https://doi.org/10.1038%2F4456
5). PMID 10548103 (https://pubmed.ncbi.nlm.nih.gov/10548103). S2CID 4428232 (https://ap
i.semanticscholar.org/CorpusID:4428232).
43. Wray Buntine (2002). Variational Extensions to EM and Multinomial PCA (http://cosco.hiit.fi/
Articles/ecml02.pdf) (PDF). Proc. European Conference on Machine Learning (ECML-02).
LNAI. Vol. 2430. pp. 23–34.
44. Eric Gaussier & Cyril Goutte (2005). Relation between PLSA and NMF and Implications (htt
ps://web.archive.org/web/20070928032454/http://eprints.pascal-network.org/archive/000009
71/01/39-gaussier.pdf) (PDF). Proc. 28th international ACM SIGIR conference on Research
and development in information retrieval (SIGIR-05). pp. 601–602. Archived from the original
(http://eprints.pascal-network.org/archive/00000971/01/39-gaussier.pdf) (PDF) on 2007-09-
28. Retrieved 2007-01-29.
45. Ron Zass and Amnon Shashua (2005). "A Unifying Approach to Hard and Probabilistic
Clustering (http://www.cs.huji.ac.il/~zass/papers/cp-iccv05.pdf)". International Conference on
Computer Vision (ICCV) Beijing, China, Oct., 2005.
46. Max Welling; et al. (2004). Exponential Family Harmoniums with an Application to
Information Retrieval (http://papers.nips.cc/paper/2672-exponential-family-harmoniums-with-
an-application-to-information-retrieval). NIPS.
47. Pentti Paatero (1999). "The Multilinear Engine: A Table-Driven, Least Squares Program for
Solving Multilinear Problems, including the n-Way Parallel Factor Analysis Model". Journal
of Computational and Graphical Statistics. 8 (4): 854–888. doi:10.2307/1390831 (https://doi.
org/10.2307%2F1390831). JSTOR 1390831 (https://www.jstor.org/stable/1390831).
48. Max Welling & Markus Weber (2001). "Positive Tensor Factorization". Pattern Recognition
Letters. 22 (12): 1255–1261. Bibcode:2001PaReL..22.1255W (https://ui.adsabs.harvard.edu/
abs/2001PaReL..22.1255W). CiteSeerX 10.1.1.21.24 (https://citeseerx.ist.psu.edu/viewdoc/s
ummary?doi=10.1.1.21.24). doi:10.1016/S0167-8655(01)00070-8 (https://doi.org/10.1016%2
FS0167-8655%2801%2900070-8).
49. Jingu Kim & Haesun Park (2012). Fast Nonnegative Tensor Factorization with an Active-set-
like Method (http://www.cc.gatech.edu/~hpark/papers/2011_paper_hpscbook_ntf.pdf) (PDF).
High-Performance Scientific Computing: Algorithms and Applications. Springer. pp. 311–
326.
50. Kenan Yilmaz; A. Taylan Cemgil & Umut Simsekli (2011). Generalized Coupled Tensor
Factorization (http://books.nips.cc/papers/files/nips24/NIPS2011_1189.pdf) (PDF). NIPS.
51. Vamsi K. Potluru; Sergey M. Plis; Morten Morup; Vince D. Calhoun & Terran Lane (2009).
Efficient Multiplicative updates for Support Vector Machines. Proceedings of the 2009 SIAM
Conference on Data Mining (SDM). pp. 1218–1229.
52. Wei Xu; Xin Liu & Yihong Gong (2003). Document clustering based on non-negative matrix
factorization (http://portal.acm.org/citation.cfm?id=860485). Proceedings of the 26th annual
international ACM SIGIR conference on Research and development in information retrieval.
New York: Association for Computing Machinery. pp. 267–273.
53. Eggert, J.; Korner, E. (2004). "Sparse coding and NMF". 2004 IEEE International Joint
Conference on Neural Networks (IEEE Cat. No.04CH37541). Vol. 4. pp. 2529–2533.
doi:10.1109/IJCNN.2004.1381036 (https://doi.org/10.1109%2FIJCNN.2004.1381036).
ISBN 978-0-7803-8359-3. S2CID 17923083 (https://api.semanticscholar.org/CorpusID:1792
3083).
54. Berné, O.; Joblin, C.; Deville, Y.; Smith, J. D.; Rapacioli, M.; Bernard, J. P.; Thomas, J.;
Reach, W.; Abergel, A. (2007-07-01). "Analysis of the emission of very small dust particles
from Spitzer spectro-imagery data using blind signal separation methods" (https://www.aand
a.org/articles/aa/abs/2007/26/aa6282-06/aa6282-06.html). Astronomy & Astrophysics. 469
(2): 575–586. doi:10.1051/0004-6361:20066282 (https://doi.org/10.1051%2F0004-6361%3A
20066282). ISSN 0004-6361 (https://www.worldcat.org/issn/0004-6361).
55. Lafrenière, David; Maroid, Christian; Doyon, René; Barman, Travis (2009). "HST/NICMOS
Detection of HR 8799 b in 1998". The Astrophysical Journal Letters. 694 (2): L148.
arXiv:0902.3247 (https://arxiv.org/abs/0902.3247). Bibcode:2009ApJ...694L.148L (https://ui.a
dsabs.harvard.edu/abs/2009ApJ...694L.148L). doi:10.1088/0004-637X/694/2/L148 (https://d
oi.org/10.1088%2F0004-637X%2F694%2F2%2FL148). S2CID 7332750 (https://api.semanti
cscholar.org/CorpusID:7332750).
56. Amara, Adam; Quanz, Sascha P. (2012). "PYNPOINT: an image processing package for
finding exoplanets". Monthly Notices of the Royal Astronomical Society. 427 (2): 948.
arXiv:1207.6637 (https://arxiv.org/abs/1207.6637). Bibcode:2012MNRAS.427..948A (https://
ui.adsabs.harvard.edu/abs/2012MNRAS.427..948A). doi:10.1111/j.1365-2966.2012.21918.x
(https://doi.org/10.1111%2Fj.1365-2966.2012.21918.x). S2CID 119200505 (https://api.sema
nticscholar.org/CorpusID:119200505).
57. Wahhaj, Zahed; Cieza, Lucas A.; Mawet, Dimitri; Yang, Bin; Canovas, Hector; de Boer,
Jozua; Casassus, Simon; Ménard, François; Schreiber, Matthias R.; Liu, Michael C.; Biller,
Beth A.; Nielsen, Eric L.; Hayward, Thomas L. (2015). "Improving signal-to-noise in the direct
imaging of exoplanets and circumstellar disks with MLOCI". Astronomy & Astrophysics. 581
(24): A24. arXiv:1502.03092 (https://arxiv.org/abs/1502.03092).
Bibcode:2015A&A...581A..24W (https://ui.adsabs.harvard.edu/abs/2015A&A...581A..24W).
doi:10.1051/0004-6361/201525837 (https://doi.org/10.1051%2F0004-6361%2F201525837).
S2CID 20174209 (https://api.semanticscholar.org/CorpusID:20174209).
58. Nielsen, Finn Årup; Balslev, Daniela; Hansen, Lars Kai (2005). "Mining the posterior
cingulate: segregation between memory and pain components" (http://orbit.dtu.dk/ws/files/39
36747/imm3661.pdf) (PDF). NeuroImage. 27 (3): 520–522.
doi:10.1016/j.neuroimage.2005.04.034 (https://doi.org/10.1016%2Fj.neuroimage.2005.04.03
4). PMID 15946864 (https://pubmed.ncbi.nlm.nih.gov/15946864). S2CID 18509039 (https://a
pi.semanticscholar.org/CorpusID:18509039).
59. Cohen, William (2005-04-04). "Enron Email Dataset" (https://www.cs.cmu.edu/~enron/).
Retrieved 2008-08-26.
60. Berry, Michael W.; Browne, Murray (2005). "Email Surveillance Using Non-negative Matrix
Factorization". Computational and Mathematical Organization Theory. 11 (3): 249–264.
doi:10.1007/s10588-005-5380-5 (https://doi.org/10.1007%2Fs10588-005-5380-5).
S2CID 16249147 (https://api.semanticscholar.org/CorpusID:16249147).
61. Nielsen, Finn Årup (2008). Clustering of scientific citations in Wikipedia (http://www2.imm.dt
u.dk/pubdb/views/publication_details.php?id=5666). Wikimania. arXiv:0805.1154 (https://arx
iv.org/abs/0805.1154).
62. Hassani, Ali; Iranmanesh, Amir; Mansouri, Najme (2019-11-12). "Text Mining using
Nonnegative Matrix Factorization and Latent Semantic Analysis". arXiv:1911.04705 (https://
arxiv.org/abs/1911.04705) [cs.LG (https://arxiv.org/archive/cs.LG)].
63. Berry, Michael W.; Browne, Murray; Langville, Amy N.; Paucac, V. Paul; Plemmonsc, Robert
J. (15 September 2007). "Algorithms and Applications for Approximate Nonnegative Matrix
Factorization". Computational Statistics & Data Analysis. 52 (1): 155–173.
doi:10.1016/j.csda.2006.11.006 (https://doi.org/10.1016%2Fj.csda.2006.11.006).
64. Yun Mao; Lawrence Saul & Jonathan M. Smith (2006). "IDES: An Internet Distance
Estimation Service for Large Networks". IEEE Journal on Selected Areas in
Communications. 24 (12): 2273–2284. CiteSeerX 10.1.1.136.3837 (https://citeseerx.ist.psu.e
du/viewdoc/summary?doi=10.1.1.136.3837). doi:10.1109/JSAC.2006.884026 (https://doi.or
g/10.1109%2FJSAC.2006.884026). S2CID 12931155 (https://api.semanticscholar.org/Corp
usID:12931155).
65. Yang Chen; Xiao Wang; Cong Shi; et al. (2011). "Phoenix: A Weight-based Network
Coordinate System Using Matrix Factorization" (https://web.archive.org/web/201111141912
20/http://www.cs.duke.edu/~ychen/Phoenix_TNSM.pdf) (PDF). IEEE Transactions on
Network and Service Management. 8 (4): 334–347. CiteSeerX 10.1.1.300.2851 (https://cites
eerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.300.2851).
doi:10.1109/tnsm.2011.110911.100079 (https://doi.org/10.1109%2Ftnsm.2011.110911.1000
79). S2CID 8079061 (https://api.semanticscholar.org/CorpusID:8079061). Archived from the
original (http://www.cs.duke.edu/~ychen/Phoenix_TNSM.pdf) (PDF) on 2011-11-14.
66. Schmidt, M.N., J. Larsen, and F.T. Hsiao. (2007). "Wind noise reduction using non-negative
sparse coding (http://orbit.dtu.dk/files/3848474/Schmidt.pdf)", Machine Learning for Signal
Processing, IEEE Workshop on, 431–436
67. Frichot E, Mathieu F, Trouillon T, Bouchard G, Francois O (2014). "Fast and efficient
estimation of individual ancestry coefficients" (https://www.ncbi.nlm.nih.gov/pmc/articles/PM
C3982712). Genetics. 196 (4): 973–983. doi:10.1534/genetics.113.160572 (https://doi.org/1
0.1534%2Fgenetics.113.160572). PMC 3982712 (https://www.ncbi.nlm.nih.gov/pmc/articles/
PMC3982712). PMID 24496008 (https://pubmed.ncbi.nlm.nih.gov/24496008).
68. Devarajan, K. (2008). "Nonnegative Matrix Factorization: An Analytical and Interpretive Tool
in Computational Biology" (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2447881). PLOS
Computational Biology. 4 (7): e1000029. Bibcode:2008PLSCB...4E0029D (https://ui.adsabs.
harvard.edu/abs/2008PLSCB...4E0029D). doi:10.1371/journal.pcbi.1000029 (https://doi.org/
10.1371%2Fjournal.pcbi.1000029). PMC 2447881 (https://www.ncbi.nlm.nih.gov/pmc/article
s/PMC2447881). PMID 18654623 (https://pubmed.ncbi.nlm.nih.gov/18654623).
69. Hyunsoo Kim & Haesun Park (2007). "Sparse non-negative matrix factorizations via
alternating non-negativity-constrained least squares for microarray data analysis" (https://do
i.org/10.1093%2Fbioinformatics%2Fbtm134). Bioinformatics. 23 (12): 1495–1502.
doi:10.1093/bioinformatics/btm134 (https://doi.org/10.1093%2Fbioinformatics%2Fbtm134).
PMID 17483501 (https://pubmed.ncbi.nlm.nih.gov/17483501).
70. Schwalbe, E. (2013). "DNA methylation profiling of medulloblastoma allows robust sub-
classification and improved outcome prediction using formalin-fixed biopsies" (https://www.n
cbi.nlm.nih.gov/pmc/articles/PMC4313078). Acta Neuropathologica. 125 (3): 359–371.
doi:10.1007/s00401-012-1077-2 (https://doi.org/10.1007%2Fs00401-012-1077-2).
PMC 4313078 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4313078). PMID 23291781
(https://pubmed.ncbi.nlm.nih.gov/23291781).
71. Alexandrov, Ludmil B.; Nik-Zainal, Serena; Wedge, David C.; Campbell, Peter J.; Stratton,
Michael R. (2013-01-31). "Deciphering signatures of mutational processes operative in
human cancer" (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3588146). Cell Reports. 3
(1): 246–259. doi:10.1016/j.celrep.2012.12.008 (https://doi.org/10.1016%2Fj.celrep.2012.12.
008). ISSN 2211-1247 (https://www.worldcat.org/issn/2211-1247). PMC 3588146 (https://ww
w.ncbi.nlm.nih.gov/pmc/articles/PMC3588146). PMID 23318258 (https://pubmed.ncbi.nlm.ni
h.gov/23318258).
72. Stein-O’Brien, Genevieve L.; Arora, Raman; Culhane, Aedin C.; Favorov, Alexander V.;
Garmire, Lana X.; Greene, Casey S.; Goff, Loyal A.; Li, Yifeng; Ngom, Aloune; Ochs, Michael
F.; Xu, Yanxun (2018-10-01). "Enter the Matrix: Factorization Uncovers Knowledge from
Omics" (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6309559). Trends in Genetics. 34
(10): 790–805. doi:10.1016/j.tig.2018.07.003 (https://doi.org/10.1016%2Fj.tig.2018.07.003).
ISSN 0168-9525 (https://www.worldcat.org/issn/0168-9525). PMC 6309559 (https://www.ncb
i.nlm.nih.gov/pmc/articles/PMC6309559). PMID 30143323 (https://pubmed.ncbi.nlm.nih.gov/
30143323).
73. Ding; Li; Peng; Park (2006). "Orthogonal nonnegative matrix t-factorizations for clustering".
Proceedings of the 12th ACM SIGKDD International Conference on Knowledge Discovery
and Data Mining: 126–135. doi:10.1145/1150402.1150420 (https://doi.org/10.1145%2F1150
402.1150420). ISBN 1595933395. S2CID 165018 (https://api.semanticscholar.org/CorpusI
D:165018).
74. Ceddia; Pinoli; Ceri; Masseroli (2020). "Matrix factorization-based technique for drug
repurposing predictions". IEEE Journal of Biomedical and Health Informatics. 24 (11): 3162–
3172. doi:10.1109/JBHI.2020.2991763 (https://doi.org/10.1109%2FJBHI.2020.2991763).
PMID 32365039 (https://pubmed.ncbi.nlm.nih.gov/32365039). S2CID 218504587 (https://ap
i.semanticscholar.org/CorpusID:218504587).
75. Pinoli; Ceddia; Ceri; Masseroli (2021). "Predicting drug synergism by means of non-
negative matrix tri-factorization". IEEE/ACM Transactions on Computational Biology and
Bioinformatics. PP (4): 1956–1967. doi:10.1109/TCBB.2021.3091814 (https://doi.org/10.110
9%2FTCBB.2021.3091814). PMID 34166199 (https://pubmed.ncbi.nlm.nih.gov/34166199).
S2CID 235634059 (https://api.semanticscholar.org/CorpusID:235634059).
76. DiPaola; Bazin; Aubry; Aurengo; Cavailloles; Herry; Kahn (1982). "Handling of dynamic
sequences in nuclear medicine". IEEE Trans Nucl Sci. 29 (4): 1310–21.
Bibcode:1982ITNS...29.1310D (https://ui.adsabs.harvard.edu/abs/1982ITNS...29.1310D).
doi:10.1109/tns.1982.4332188 (https://doi.org/10.1109%2Ftns.1982.4332188).
S2CID 37186516 (https://api.semanticscholar.org/CorpusID:37186516).
77. Sitek; Gullberg; Huesman (2002). "Correction for ambiguous solutions in factor analysis
using a penalized least squares objective". IEEE Trans Med Imaging. 21 (3): 216–25.
doi:10.1109/42.996340 (https://doi.org/10.1109%2F42.996340). PMID 11989846 (https://pub
med.ncbi.nlm.nih.gov/11989846). S2CID 6553527 (https://api.semanticscholar.org/CorpusI
D:6553527).
78. Boutchko; Mitra; Baker; Jagust; Gullberg (2015). "Clustering Initiated Factor Analysis (CIFA)
Application for Tissue Classification in Dynamic Brain PET" (https://www.ncbi.nlm.nih.gov/p
mc/articles/PMC4640278). Journal of Cerebral Blood Flow and Metabolism. 35 (7): 1104–
11. doi:10.1038/jcbfm.2015.69 (https://doi.org/10.1038%2Fjcbfm.2015.69). PMC 4640278 (ht
tps://www.ncbi.nlm.nih.gov/pmc/articles/PMC4640278). PMID 25899294 (https://pubmed.nc
bi.nlm.nih.gov/25899294).
79. Abdalah; Boutchko; Mitra; Gullberg (2015). "Reconstruction of 4-D Dynamic SPECT Images
From Inconsistent Projections Using a Spline Initialized FADS Algorithm (SIFADS)" (https://e
scholarship.org/uc/item/0b95c190). IEEE Trans Med Imaging. 34 (1): 216–18.
doi:10.1109/TMI.2014.2352033 (https://doi.org/10.1109%2FTMI.2014.2352033).
PMID 25167546 (https://pubmed.ncbi.nlm.nih.gov/25167546). S2CID 11060831 (https://api.s
emanticscholar.org/CorpusID:11060831).
80. C. Boutsidis & E. Gallopoulos (2008). "SVD based initialization: A head start for nonnegative
matrix factorization". Pattern Recognition. 41 (4): 1350–1362.
Bibcode:2008PatRe..41.1350B (https://ui.adsabs.harvard.edu/abs/2008PatRe..41.1350B).
CiteSeerX 10.1.1.137.8281 (https://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.137.
8281). doi:10.1016/j.patcog.2007.09.010 (https://doi.org/10.1016%2Fj.patcog.2007.09.010).
81. Chao Liu; Hung-chih Yang; Jinliang Fan; Li-Wei He & Yi-Min Wang (2010). "Distributed
Nonnegative Matrix Factorization for Web-Scale Dyadic Data Analysis on MapReduce" (htt
p://research.microsoft.com/pubs/119077/DNMF.pdf) (PDF). Proceedings of the 19th
International World Wide Web Conference.
82. Jiangtao Yin; Lixin Gao & Zhongfei (Mark) Zhang (2014). "Scalable Nonnegative Matrix
Factorization with Block-wise Updates" (http://rio.ecs.umass.edu/mnilpub/papers/ecmlpkdd2
014-yin.pdf) (PDF). Proceedings of the European Conference on Machine Learning and
Principles and Practice of Knowledge Discovery in Databases.
83. "Apache Mahout" (https://mahout.apache.org/). mahout.apache.org. Retrieved 2019-12-14.
84. Dong Wang; Ravichander Vipperla; Nick Evans; Thomas Fang Zheng (2013). "Online Non-
Negative Convolutive Pattern Learning for Speech Signals" (https://web.archive.org/web/20
150419072552/http://cslt.riit.tsinghua.edu.cn:8081/homepages/wangd/public/pdf/cnsc-tsp.pd
f) (PDF). IEEE Transactions on Signal Processing. 61 (1): 44–56.
Bibcode:2013ITSP...61...44W (https://ui.adsabs.harvard.edu/abs/2013ITSP...61...44W).
CiteSeerX 10.1.1.707.7348 (https://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.707.
7348). doi:10.1109/tsp.2012.2222381 (https://doi.org/10.1109%2Ftsp.2012.2222381).
S2CID 12530378 (https://api.semanticscholar.org/CorpusID:12530378). Archived from the
original (http://cslt.riit.tsinghua.edu.cn:8081/homepages/wangd/public/pdf/cnsc-tsp.pdf)
(PDF) on 2015-04-19. Retrieved 2015-04-19.
85. Xiangnan He; Min-Yen Kan; Peichu Xie & Xiao Chen (2014). "Comment-based Multi-View
Clustering of Web 2.0 Items" (https://web.archive.org/web/20150402103346/http://www.com
p.nus.edu.sg/~xiangnan/files/www2014-he.pdf) (PDF). Proceedings of the 23rd International
World Wide Web Conference. Archived from the original (http://www.comp.nus.edu.sg/~xian
gnan/files/www2014-he.pdf) (PDF) on 2015-04-02. Retrieved 2015-03-22.
86. Jialu Liu; Chi Wang; Jing Gao & Jiawei Han (2013). Multi-View Clustering via Joint
Nonnegative Matrix Factorization (http://jialu.cs.illinois.edu/paper/sdm2013-liu.pdf) (PDF).
Proceedings of SIAM Data Mining Conference. pp. 252–260. CiteSeerX 10.1.1.301.1771 (htt
ps://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.301.1771).
doi:10.1137/1.9781611972832.28 (https://doi.org/10.1137%2F1.9781611972832.28).
ISBN 978-1-61197-262-7. S2CID 4968 (https://api.semanticscholar.org/CorpusID:4968).
87. Chistikov, Dmitry; Kiefer, Stefan; Marušić, Ines; Shirmohammadi, Mahsa; Worrell, James
(2016-05-22). "Nonnegative Matrix Factorization Requires Irrationality". arXiv:1605.06848 (h
ttps://arxiv.org/abs/1605.06848) [cs.CC (https://arxiv.org/archive/cs.CC)].

Others
J. Shen; G. W. Israël (1989). "A receptor model using a specific non-negative transformation
technique for ambient aerosol". Atmospheric Environment. 23 (10): 2289–2298.
Bibcode:1989AtmEn..23.2289S (https://ui.adsabs.harvard.edu/abs/1989AtmEn..23.2289S).
doi:10.1016/0004-6981(89)90190-X (https://doi.org/10.1016%2F0004-6981%2889%299019
0-X).
Pentti Paatero (1997). "Least squares formulation of robust non-negative factor analysis".
Chemometrics and Intelligent Laboratory Systems. 37 (1): 23–35. doi:10.1016/S0169-
7439(96)00044-5 (https://doi.org/10.1016%2FS0169-7439%2896%2900044-5).
Raul Kompass (2007). "A Generalized Divergence Measure for Nonnegative Matrix
Factorization". Neural Computation. 19 (3): 780–791. doi:10.1162/neco.2007.19.3.780 (http
s://doi.org/10.1162%2Fneco.2007.19.3.780). PMID 17298233 (https://pubmed.ncbi.nlm.nih.g
ov/17298233). S2CID 5337451 (https://api.semanticscholar.org/CorpusID:5337451).
Liu, W.X.; Zheng, N.N. & You, Q.B. (2006). "Nonnegative Matrix Factorization and its
applications in pattern recognition". Chinese Science Bulletin. 51 (17–18): 7–18.
 Bibcode:2006ChSBu..51....7L (https://ui.adsabs.harvard.edu/abs/2006ChSBu..51....7L).
doi:10.1007/s11434-005-1109-6 (https://doi.org/10.1007%2Fs11434-005-1109-6).
S2CID 15445516 (https://api.semanticscholar.org/CorpusID:15445516).
Ngoc-Diep Ho; Paul Van Dooren & Vincent Blondel (2008). "Descent Methods for
Nonnegative Matrix Factorization". arXiv:0801.3199 (https://arxiv.org/abs/0801.3199) [cs.NA
(https://arxiv.org/archive/cs.NA)].
Andrzej Cichocki; Rafal Zdunek & Shun-ichi Amari (2008). "Nonnegative Matrix and Tensor
Factorization". IEEE Signal Processing Magazine. 25 (1): 142–145.
Bibcode:2008ISPM...25R.142C (https://ui.adsabs.harvard.edu/abs/2008ISPM...25R.142C).
doi:10.1109/MSP.2008.4408452 (https://doi.org/10.1109%2FMSP.2008.4408452).
S2CID 9997603 (https://api.semanticscholar.org/CorpusID:9997603).
Cédric Févotte; Nancy Bertin & Jean-Louis Durrieu (2009). "Nonnegative Matrix
Factorization with the Itakura-Saito Divergence: With Application to Music Analysis". Neural
Computation. 21 (3): 793–830. doi:10.1162/neco.2008.04-08-771 (https://doi.org/10.1162%2
Fneco.2008.04-08-771). PMID 18785855 (https://pubmed.ncbi.nlm.nih.gov/18785855).
S2CID 13208611 (https://api.semanticscholar.org/CorpusID:13208611).
Ali Taylan Cemgil (2009). "Bayesian Inference for Nonnegative Matrix Factorisation Models"
(https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2688815). Computational Intelligence and
Neuroscience. 2009 (2): 1–17. doi:10.1155/2009/785152 (https://doi.org/10.1155%2F2009%
2F785152). PMC 2688815 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2688815).
PMID 19536273 (https://pubmed.ncbi.nlm.nih.gov/19536273).
Andrzej Cichocki, Morten Mrup, et al.: "Advances in Nonnegative Matrix and Tensor
Factorization", Hindawi Publishing Corporation, ISBN 978-9774540455 (2008).
Andrzej Cichocki, Rafal Zdunek, Anh Huy Phan and Shun-ichi Amari: "Nonnegative Matrix
and Tensor Factorizations: Applications to Exploratory Multi-way Data Analysis and Blind
Source Separation", Wiley, ISBN 978-0470746660 (2009).
Andri Mirzal: "Nonnegative Matrix Factorizations for Clustering and LSI: Theory and
Programming", LAP LAMBERT Academic Publishing, ISBN 978-3844324891 (2011).
Yong Xiang: "Blind Source Separation: Dependent Component Analysis", Springer,
ISBN 978-9812872265 (2014).
Ganesh R. Naik(Ed.): "Non-negative Matrix Factorization Techniques: Advances in Theory
and Applications", Springer, ISBN 978-3662517000 (2016).
Julian Becker: "Nonnegative Matrix Factorization with Adaptive Elements for Monaural
Audio Source Separation: 1 ", Shaker Verlag GmbH, Germany, ISBN 978-3844048148
(2016).
Jen-Tzung Chien: "Source Separation and Machine Learning", Academic Press, ISBN 978-
0128177969 (2018).
Shoji Makino(Ed.): "Audio Source Separation", Springer, ISBN 978-3030103033 (2019).
Nicolas Gillis: "Nonnegative Matrix Factorization", SIAM, ISBN 978-1-611976-40-3 (2020).

Retrieved from "https://en.wikipedia.org/w/index.php?title=Non-negative_matrix_factorization&oldid=1156221167"