Extraction of Food Consumption Systems by Nonnegative Matrix Factorization (NMF) for the Assessment of Food Choices

Autor: Max Feinberg, Mélanie Zetlaoui, Philippe Verger, Stéphan Clémençon
Přispěvatelé: Modélisation aléatoire de Paris X (MODAL'X), Université Paris Nanterre (UPN), UAR 1306 Direction du Système d'Information-Unité d'Appui, Institut National de la Recherche Agronomique (INRA)-Direction du Système d'Information (DSI)-Direction du Système d'Information-Unité d'Appui (DSI-UA ), WHO, Laboratoire Traitement et Communication de l'Information (LTCI), Télécom ParisTech-Institut Mines-Télécom [Paris] (IMT), Département Images, Données, Signal (IDS), Télécom ParisTech, Méthodologies d'Analyse de Risque Alimentaire (MET@RISK), Institut National de la Recherche Agronomique (INRA), Télécom ParisTech-Institut Mines-Télécom [Paris] (IMT)-Centre National de la Recherche Scientifique (CNRS), This research was supported by the project 'Tangerine' from Agence Nationale de la Recherche. It was stimulated by very interesting discussions with Drs C. Fevotte and O. Dikmen., ANR-06-BLAN-0194,TAMIS,Adaptation, tests multiples, ranking et applications(2006), Institut Mines-Télécom [Paris] (IMT)-Télécom Paris
Jazyk: angličtina
Rok vydání: 2011
Předmět:
[SDV.SA]Life Sciences [q-bio]/Agricultural sciences
sparse data
030309 nutrition & dietetics
food consumption patterns
Appetite
computer.software_genre
Choice Behavior
Pattern Recognition
Automated
[STAT.ML]Statistics [stat]/Machine Learning [stat.ML]
[MATH.MATH-ST]Mathematics [math]/Statistics [math.ST]
Non-negative Matrix Factorization (NMF)
Dimension (data warehouse)
[MATH]Mathematics [math]
life sciences and biomedicine
ComputingMilieux_MISCELLANEOUS
Mathematics
dimensionality reduction
2. Zero hunger
0303 health sciences
mathematics
Applied Mathematics
04 agricultural and veterinary sciences
General Medicine
040401 food science
Data Interpretation
Statistical
Data mining
General Agricultural and Biological Sciences
[STAT.ME]Statistics [stat]/Methodology [stat.ME]
Algorithms
Statistics and Probability
Machine learning
External Data Representation
General Biochemistry
Genetics and Molecular Biology
Non-negative matrix factorization
Food Preferences
03 medical and health sciences
0404 agricultural biotechnology
Humans
Cluster analysis
Representation (mathematics)
Sparse matrix
Consumption (economics)
General Immunology and Microbiology
business.industry
Dimensionality reduction
[STAT.TH]Statistics [stat]/Statistics Theory [stat.TH]
mathematical and computational biology
nonnegative matrix factorization (NMF)
[MATH.MATH-PR]Mathematics [math]/Probability [math.PR]
Artificial intelligence
business
computer
NMF contribution clustering
Zdroj: Biometrics
Biometrics, Wiley, 2011, 67 (4), pp.1647-1658. ⟨10.1111/j.1541-0420.2011.01588.x⟩
Biometrics, 2011, 67 (4), pp.1647-1658. ⟨10.1111/j.1541-0420.2011.01588.x⟩
ISSN: 0006-341X
1541-0420
Popis: In Western countries where food supply is satisfactory, consumers organize their diets around a large combination of foods. It is the purpose of this paper to examine how recent nonnegative matrix factorization (NMF) techniques can be applied to food consumption data in order to understand these combinations. Such data are nonnegative by nature and of high dimension. The NMF model provides a representation of consumption data through latent vectors with nonnegative coefficients, we call consumption systems, in a small number. As the NMF approach may encourage sparsity of the data representation produced, the resulting consumption systems are easily interpretable. Beyond the illustration of its properties we provide through a simple simulation result, the NMF method is applied to data issued from a french consumption survey. The numerical results thus obtained are displayed and thoroughly discussed. A clustering based on the k-means method is also achieved in the resulting latent consumption space, in order to recover food consumption patterns easily usable for nutritionists.
Databáze: OpenAIRE
Externí odkaz: