Éditeur(s) : HAL CCSD
Résumé : International audience
EmailPrintRequest PermissionsThe emergence of new communication means such as online newspapers and exchange and sharing websites allow us to go further in the understanding of diffusion phenomena. Indeed, these public discussion areas are now firmly established in our societies and are known to be strong sensors of both human behaviors and collective feelings. In this paper we focus on diffusion phenomena that occur on Twitter and we propose a set of measures that aims to characterize globally and locally the processes. Our objective is to identify what are the conditions in which a person decides to forward the information. Our measures have been used to study two events occurred in January 2015: the presentation of Microsoft HoloLens, a new augmented reality headset, and the political election in Greece. The results obtained show a strong heterogeneity in the individual behaviours involved in the diffusion process. Our approach has been implemented into a graphical tool which is able to conduct real time analysis on any kind of topics occurring on Twitter.
IEEE International Conference on Research Challenges in Information Science (RCIS)
Athens, Greece

hal-01312382
https://hal.univ-antilles.fr/hal-01312382
DOI : 10.1109/RCIS.2015.7128894

Éditeur(s) : HAL CCSD
Résumé : International audience
This paper shows the information criteria (IC) performances in feature selection framework. Feature selection aims to select a representative subset among a wide set of features. We apply this approach to classify an hand segmented image. The performance is tested using various feature selection schemes (SFS, SBS, SFFS and SBFS) to select the candidate subsets. The accuracy of the approach is based on a good quality of the joint probability density approximation of the combined features. They are obtained using histogram optimized thanks to the adaptive arithmetic coding principles. Our approach is tested on different reference data. The subsets quality is evaluated using correct classification rate computed on multiple classifiers. Results show stability and convergence properties of this tool and its ability to select representative subsets (in the sense that the subset of feature is a good characterization of the classes in which the data belong). Information Criteria could be used for feature selection as a good alternative to other criteria.
CISP Proceedings
Shangay, China

hal-00634749
https://hal.archives-ouvertes.fr/hal-00634749

Éditeur(s) : HAL CCSD
Résumé : International audience
Colloque Télédétection de Outils Mathématiques pour le Développement Durable (TOMaADD)
Cayenne, French Guiana

hal-01313396
https://hal.archives-ouvertes.fr/hal-01313396

Éditeur(s) : HAL CCSD
Douzième journées du groupe MODE
Le Havre, France

hal-00508465
https://hal.archives-ouvertes.fr/hal-00508465

Éditeur(s) : HAL CCSD Heldermann
Résumé : International audience
Journal of Convex Analysis

hal-00507803
https://hal.archives-ouvertes.fr/hal-00507803

Éditeur(s) : HAL CCSD
Résumé : International audience
Comprendre la Guyane d'aujourd'hui. Un département français dans la région des Guyanes,

hal-01656084
https://hal.archives-ouvertes.fr/hal-01656084
https://hal.archives-ouvertes.fr/hal-01656084/document
https://hal.archives-ouvertes.fr/hal-01656084/file/I.%20DUBOST%20TERRITORIALITE%20DES%20CHINOIS%20ET%20DES%20LIBANAIS%20GUYANAIS.pdf

Éditeur(s) : HAL CCSD Institute of Electrical and Electronics Engineers
Résumé : International audience
This paper presents our response to the first interna- tional challenge on Facial Emotion Recognition and Analysis. We propose to combine different types of features to automatically detect Action Units in facial images. We use one multi-kernel SVM for each Action Unit we want to detect. The first kernel matrix is computed using Local Gabor Binary Pattern histograms and a histogram intersection kernel. The second kernel matrix is computed from AAM coefficients and an RBF kernel. During the training step, we combine these two types of features using the recently proposed SimpleMKL algorithm. SVM outputs are then averaged to exploit temporal information in the sequence. To eval- uate our system, we perform deep experimentations on several key issues: influence of features and kernel function in histogram- based SVM approaches, influence of spatially-independent in- formation versus geometric local appearance information and benefits of combining both, sensitivity to training data and interest of temporal context adaptation. We also compare our results to those of the other participants and try to explain why our method had the best performance during the FERA challenge.
ISSN: 1083-4419

hal-00731864
http://hal.upmc.fr/hal-00731864
http://hal.upmc.fr/hal-00731864/document
http://hal.upmc.fr/hal-00731864/file/senechal12tsmc_preprint.pdf
DOI : 10.1109/TSMCB.2012.2193567

Éditeur(s) : HAL CCSD Elsevier
Résumé : International audience
A set-valued upper semi-continuous map is called an Rδ -map if each of its values is an Rδ -set (we recall that an Rδ -set is a space that can be represented as the intersection of a decreasing sequence of compact AR-spaces). We prove that a compact set-valued map of an AR-space into itself has a fixed point provided it can be factorized by an arbitrary finite number of Rδ -maps through ANR-spaces. This fact is a consequence of a more general result which is the main goal of this note. The proof relies on a refinement of the approximation technique and does not make use of homological tools.
ISSN: 0166-8641

hal-00699222
https://hal.archives-ouvertes.fr/hal-00699222

Éditeur(s) : HAL CCSD
Séminaire du Laboratoire Lamia
Pointe-à-Pitre, Guadeloupe

hal-00606660
https://hal.archives-ouvertes.fr/hal-00606660

Éditeur(s) : HAL CCSD Elsevier
Résumé : International audience
Journal of Mathematical Analysis and applications

hal-00624139
https://hal.archives-ouvertes.fr/hal-00624139

Éditeur(s) : HAL CCSD
Résumé : International audience
Workshop Satellite Meetings ICIAM'2011
Vancouver, Canada

hal-00770279
https://hal.univ-antilles.fr/hal-00770279

Éditeur(s) : HAL CCSD Elsevier
Résumé : International audience
ISSN: 0003-4975

hal-01563338
http://hal.univ-brest.fr/hal-01563338
DOI : 10.1016/j.athoracsur.2016.03.093
PUBMED : 27262916

Éditeur(s) : HAL CCSD
Résumé : International audience
This paper is concerned with a null controllability problem for nonlinear parabolic equations appearing in mathematical biology. We consider a system of two equations with a nonlocal nonlinearity, in a bounded domain Ω of ℝN0, N0 being a positive integer sufficiently small. Such models can be used in the context of dynamics of biological systems to describe the migration of population (of bacteria in a container). Our aim is to find a control common to both equations, which solves the null controllabillity of the studied system under constraints on the state. A similar control problem has recently been addressed for the Laplace operator, so the present paper extends those results.
2017 Proceedings of the Conference on Control and its Applications
Pittsburgh, United States

Droits : http://hal.archives-ouvertes.fr/licences/copyright/
hal-01567339
https://hal.archives-ouvertes.fr/hal-01567339
DOI : 10.1137/1.9781611975024.3

Éditeur(s) : HAL CCSD AIMS
Résumé : International audience
Mathematical Biosciences and Engineering

hal-01313457
https://hal.archives-ouvertes.fr/hal-01313457

Éditeur(s) : HAL CCSD
Résumé : 6 pages, 4 figures. In ICML 2013 Workshop on Machine Learning for Bioacoustics
International audience
The work presented in this paper is part of a global framework which long term goal is to design a wireless sensor network able to support the observation of a population of endangered birds. We present the first stage for which we have conducted a knowledge discovery approach on a sample of acoustical data. We use MFCC features extracted from bird songs and we exploit two knowledge discovery techniques. One that relies on clustering-based approaches, that highlights the homogeneity in the songs of the species. The other, based on predictive modeling, that demonstrates the good performances of various machine learning techniques for the identification process. The knowledge elicited provides promising results to consider a widespread study and to elicit guidelines for designing a first version of the automatic approach for data collection based on acoustic sensors.
ICML 2013 Workshop on Machine Learning for Bioacoustics
Atlanta, United States

hal-00840741
https://hal.archives-ouvertes.fr/hal-00840741
ARXIV : 1306.5349

Éditeur(s) : HAL CCSD Hong Kong : World Academic Publishing
Résumé : International audience
In automatic image analysis, when using natural images, it is important to select the elements in a scene that can lead to correct characterization. Extraneous elements can result in poor values in the parameter evaluation for recognition. Irrelevant elements also increase the processing time, because they can be used incorrectly by the algorithms for characterization or identification. The goal of this paper is to offer identification algorithms and characterizations that will allow effective recognition. We base our procedure on the photographer focus areas to identify those relevant elements in a scene. In this work, we propose a procedure to select the relevant shape from natural images. The treatment uses focal depth with edge detectors. The resulting points are combined with a region partition to obtain relevant shapes.
ISSN: 2309-0650

hal-01278550
https://hal.archives-ouvertes.fr/hal-01278550

Éditeur(s) : HAL CCSD Springer Verlag
Résumé : International audience
ISSN: 0035-5038

hal-00515752
https://hal.archives-ouvertes.fr/hal-00515752

Éditeur(s) : HAL CCSD Springer Optimization and Its Applications, Vol. 49
Résumé : International audience
Fixed-Point Algorithms for Inverse Problems in Science and Engineering

hal-00756766
https://hal.archives-ouvertes.fr/hal-00756766

Éditeur(s) : HAL CCSD
Résumé : In this paper we define the $p$-density of a finite subset $Dsubsetma{N}^r$, and show that it gives a good lower bound for the $p$-adic valuation of exponential sums over finite fields of characteristic $p$. We also give an application: when $r=1$, the $p$-density is the first slope of the generic Newton polygon of the family of Artin-Schreier curves associated to polynomials with their exponents in $D$.
https://hal.archives-ouvertes.fr/hal-00551478

hal-00551478
https://hal.archives-ouvertes.fr/hal-00551478
ARXIV : 0812.3382

Éditeur(s) : HAL CCSD RNTI
Résumé : International audience
De nombreuses méthodes ont été proposées pour extraire des clus-ters des réseaux sociaux. Si un travail important est aujourd’hui mené sur laconception de méthodes innovantes capables de rechercher des clusters de na-ture différente, la plupart des approches font l’hypothèse de réseaux statiques.L’une des récentes méthodes concerne notamment la recherche de liens concep-tuels. Il s’agit d’une nouvelle approche de clustering de liens, qui exploite à lafois la structure du réseau et les attributs des noeuds dans le but d’identifier desliens fréquents entre des groupes de noeuds au sein desquels les noeuds par-tagent des attributs communs. Dans ce travail, nous nous intéressons au suivides liens conceptuels dans des réseaux dynamiques, c’est-à-dire des réseaux quiconnaissent des changements structurels importants. Nous cherchons en parti-culier à comprendre comment les liens conceptuels se forment et évoluent aucours du développement du réseau. Pour ce faire, nous proposons un ensemblede mesures qui visent à capturer des comportements caractérisant l’évolutionde ces clusters. Notre approche est ainsi utilisée pour comprendre l’évolutiondes liens conceptuels extraits sur deux réseaux réels : un réseau de co-auteursd’articles scientifiques et un réseau de communications mobiles. Les résultatsobtenus permettent de mettre en lumière des tendances significatives dans l’évo-lution des clusters sur ces deux réseaux.
Extraction et Gestion des Connaissances (EGC) 2017. Revue des Nouvelles Technologies de l'Information vol.RNTI-E-33
Grenoble, France

hal-01514245
https://hal.archives-ouvertes.fr/hal-01514245