Part-Based Data Analysis with Masked Non-negative Matrix Factorization

2014 ◽  
pp. 440-454 ◽  
Author(s):  
Gabriella Casalino ◽  
Nicoletta Del Buono ◽  
Corrado Mencar
Keyword(s):  
Data Analysis ◽  
Matrix Factorization ◽  
Non Negative Matrix Factorization

scholarly journals Application of non-negative matrix factorization to multispectral FLIM data analysis

2012 ◽  
Vol 3 (9) ◽  
pp. 2244 ◽  
Author(s):  
Paritosh Pande ◽  
Brian E. Applegate ◽  
Javier A. Jo
Keyword(s):  
Data Analysis ◽  
Matrix Factorization ◽  
Non Negative Matrix Factorization

scholarly journals Raman imaging reveals in-situ microchemistry of cuticle and epidermis of spruce needles

Plant Methods ◽  
2021 ◽  
Vol 17 (1) ◽  
Author(s):  
Nadia Sasani ◽  
Peter Bock ◽  
Martin Felhofer ◽  
Notburga Gierlinger
Keyword(s):  
Data Analysis ◽  
Matrix Factorization ◽  
Abiotic Stresses ◽  
Protective Layer ◽  
Raman Imaging ◽  
Chemical Extraction ◽  
Chemical Heterogeneity ◽  
Coumaric Acid ◽  
Biotic And Abiotic Stresses ◽  
Non Negative Matrix Factorization

Abstract Background The cuticle is a protective layer playing an important role in plant defense against biotic and abiotic stresses. So far cuticle structure and chemistry was mainly studied by electron microscopy and chemical extraction. Thus, analysing composition involved sample destruction and the link between chemistry and microstructure remained unclear. In the last decade, Raman imaging showed high potential to link plant anatomical structure with microchemistry and to give insights into orientation of molecules. In this study, we use Raman imaging and polarization experiments to study the native cuticle and epidermal layer of needles of Norway spruce, one of the economically most important trees in Europe. The acquired hyperspectral dataset is the basis to image the chemical heterogeneity using univariate (band integration) as well as multivariate data analysis (cluster analysis and non-negative matrix factorization). Results Confocal Raman microscopy probes the cuticle together with the underlying epidermis in the native state and tracks aromatics, lipids, carbohydrates and minerals with a spatial resolution of 300 nm. All three data analysis approaches distinguish a waxy, crystalline layer on top, in which aliphatic chains and coumaric acid are aligned perpendicular to the surface. Also in the lipidic amorphous cuticle beneath, strong signals of coumaric acid and flavonoids are detected. Even the unmixing algorithm results in mixed endmember spectra and confirms that lipids co-locate with aromatics. The underlying epidermal cell walls are devoid of lipids but show strong aromatic Raman bands. Especially the upper periclinal thicker cell wall is impregnated with aromatics. At the interface between epidermis and cuticle Calcium oxalate crystals are detected in a layer-like fashion. Non-negative matrix factorization gives the purest component spectra, thus the best match with reference spectra and by this promotes band assignments and interpretation of the visualized chemical heterogeneity. Conclusions Results sharpen our view about the cuticle as the outermost layer of plants and highlight the aromatic impregnation throughout. In the future, developmental studies tracking lipid and aromatic pathways might give new insights into cuticle formation and comparative studies might deepen our understanding why some trees and their needle and leaf surfaces are more resistant to biotic and abiotic stresses than others.

scholarly journals Microbiome Data Analysis by Symmetric Non-negative Matrix Factorization With Local and Global Regularization

2021 ◽  
Vol 8 ◽  
Author(s):  
Junmin Zhao ◽  
Yuanyuan Ma ◽  
Lifang Liu
Keyword(s):  
Data Analysis ◽  
Matrix Factorization ◽  
Original Data ◽  
Biological Data ◽  
Fine Grained ◽  
Laplacian Graph ◽  
Fine Grained Structure ◽  
Microbiome Data ◽  
Global And Local ◽  
Non Negative Matrix Factorization

A network is an efficient tool to organize complicated data. The Laplacian graph has attracted more and more attention for its good properties and has been applied to many tasks including clustering, feature selection, and so on. Recently, studies have indicated that though the Laplacian graph can capture the global information of data, it lacks the power to capture fine-grained structure inherent in network. In contrast, a Vicus matrix can make full use of local topological information from the data. Given this consideration, in this paper we simultaneously introduce Laplacian and Vicus graphs into a symmetric non-negative matrix factorization framework (LVSNMF) to seek and exploit the global and local structure patterns that inherent in the original data. Extensive experiments are conducted on three real datasets (cancer, cell populations, and microbiome data). The experimental results show the proposed LVSNMF algorithm significantly outperforms other competing algorithms, suggesting its potential in biological data analysis.

scholarly journals Imaging data analysis using non-negative matrix factorization

2021 ◽  
Author(s):  
Toru Aonishi ◽  
Ryoichi Maruyama ◽  
Tsubasa Ito ◽  
Hiroyoshi Miyakawa ◽  
Masanori Murayama ◽  
...  
Keyword(s):  
Data Analysis ◽  
Matrix Factorization ◽  
Imaging Data ◽  
Non Negative Matrix Factorization

Instantaneous Versus Convolutive Non-Negative Matrix Factorization

2010 ◽  
pp. 353-370 ◽  
Author(s):  
Wenwu Wang
Keyword(s):  
Machine Learning ◽  
Performance Evaluation ◽  
Data Analysis ◽  
Matrix Factorization ◽  
Temporal Structure ◽  
Low Rank ◽  
Basis Matrix ◽  
And Performance ◽  
Separation Problems ◽  
Non Negative Matrix Factorization

Non-negative matrix factorization (NMF) is an emerging technique for data analysis and machine learning, which aims to find low-rank representations for non-negative data. Early works in NMF are mainly based on the instantaneous model, i.e. using a single basis matrix to represent the data. Recent works have shown that the instantaneous model may not be satisfactory for many audio application tasks. The convolutive NMF model, which has an advantage of revealing the temporal structure possessed by many signals, has been proposed. This chapter intends to provide a brief overview of the models and algorithms for both the instantaneous and the convolutive NMF, with a focus on the theoretical analysis and performance evaluation of the convolutive NMF algorithms, and their applications to audio pattern separation problems.

Novel Algorithm for Non-Negative Matrix Factorization

2015 ◽  
Vol 11 (02) ◽  
pp. 121-133 ◽  
Author(s):  
Tran Dang Hien ◽  
Do Van Tuan ◽  
Pham Van At ◽  
Le Hung Son
Keyword(s):  
Data Analysis ◽  
Minimization Problem ◽  
Matrix Factorization ◽  
Wide Spectrum ◽  
Popular Approach ◽  
Computational Speed ◽  
Potential Applications ◽  
Multiplicative Update ◽  
Novel Algorithm ◽  
Non Negative Matrix Factorization

Non-negative matrix factorization (NMF) is an emerging technique with a wide spectrum of potential applications in data analysis. Mathematically, NMF can be formulated as a minimization problem with non-negative constraints. This problem attracts much attention from researchers for theoretical reasons and for potential applications. Currently, the most popular approach to solve NMF is the multiplicative update algorithm proposed by Lee and Seung. In this paper, we propose an additive update algorithm that has a faster computational speed than Lee and Seung's multiplicative update algorithm.

Sign in / Sign up

Export Citation Format

Share Document