Éditeur(s) : HAL CCSD
Résumé : Works presented in this report were related to the extraction of interesting models in data mining and in the context of information system modeling.
Les travaux présentés dans ce mémoire, ont été développés sur le thème de la découverte de motifs intéressants à travers la fouille de données et mis en oeuvre dans le cadre de la conception de systèmes d'information. Ils sont essentiellement consacrés aux problèmes soulevés par l'étape de fouille pour la découverte de modèles et de motifs fréquents. Ils sont à la fois d'ordre méthodologique et applicatif.
https://tel.archives-ouvertes.fr/tel-01059407

tel-01059407
https://tel.archives-ouvertes.fr/tel-01059407
https://tel.archives-ouvertes.fr/tel-01059407/document
https://tel.archives-ouvertes.fr/tel-01059407/file/HDR.pdf

Éditeur(s) : HAL CCSD
Résumé : International audience
Geographic Information Systems (GIS) help addressing geographical and environmental issues by providing information about a region or a city as a set of maps (layers), each one displaying information about a given theme like roads, vegetation, tourist spots or museums for instance. By combining different layers on a region, one can associate a given area to characteristics from their related themes. Indeed, the information from two or more layers might be merged and then transformed into a new layer as defined in map "algebra". When a theme vocabulary is organized as a taxonomy with concepts linked by is-a relationships, there are different ways to annotate an area with a concept depending on the level selected into the layer taxonomy. In this paper, we present an heuristic- based approach for an optimal merging of such layers in a GIS. Our goal is to generate new layers which sum up information from several themes in a most useful way. Two optimization criteria are considered, the average size of resulting areas and the average informative value of their resulting annotation. We demonstrate the validity of the proposed solution, firstly, on a formal example, and then, on a real world application.
Proceedings of the 4th International Conference on Advanced Geographic Information Systems, Applications, and Services
Valencia, Spain

hal-00679555
https://hal.archives-ouvertes.fr/hal-00679555

Éditeur(s) : HAL CCSD
Résumé : International audience
This paper focuses on the study of the behaviour of a singular classifier system, the Adapted Pittsburgh Classifier System (A.P.C.S), on environments containing aliasing situations. Maze type environments are often used in reinforcement learning literature to assess the performances of learning methods when facing problems containing non Markovian situations. Those situations are often encountered when performing reinforcement learning on aliased data samples issued from meteorological simulations.
International Journal of Artificial Intelligence

hal-00520611
https://hal.archives-ouvertes.fr/hal-00520611

Éditeur(s) : HAL CCSD
Résumé : International audience
Semi-volatile organic compounds (SVOCs) partition indoors between the gas phase, airborne particles, settled dust, and other surfaces. Unknown concentrations of SVOCs in the gas phase (Cg) can be predicted from their measured concentrations in airborne particles. In previous studies, the prediction of Cg depended largely on choosing a specific equation for the calculation of the particle/gas partition coefficient. Moreover, the prediction of Cg is frequently performed at a reference temperature rather than the real indoor temperature. In this paper, a probabilistic approach based on Monte Carlo simulation was developed to predict the distribution of SVOCs' Cg from their concentrations in airborne particles at the target indoor temperature. Moreover, the distribution of the particle/gas partition coefficient of each SVOC at the target temperature was used. The approach was validated using two measured datasets in the literature: the predicted Cg from concentrations measured in airborne particles and the measured Cg were generally of the same order of magnitude. The distributions of the Cg of 66 SVOCs in the French housing stock were then predicted. The SVOCs with the highest median Cg, ranging from 1 ng/m3 to > 100 ng/m3, included 8 phthalates (DEP, DiBP, DBP, DEHP, BBP, DMP, DiNP, and DMEP), 4 polycyclic aromatic hydrocarbons (fluorene, phenanthrene, fluoranthene, and anthracene), 2 alkylphenols (4-tert-butylphenol and 4-tert-octylphenol), 2 synthetic musks (galaxolide and tonalide), tributyl phosphate, and heptachlor. The nationwide, representative, predicted Cg values of SVOCs are frequently of the same order of magnitude in Europe and North America, whereas these Cg values in Chinese and Indian dwellings and the Cg of polybrominated diphenyl ethers in U.S. dwellings are generally higher. © 2016 Elsevier B.V.
Science of the Total Environment

hal-01467547
https://hal-univ-rennes1.archives-ouvertes.fr/hal-01467547
DOI : 10.1016/j.scitotenv.2016.10.074
PUBMED : 27788447

Éditeur(s) : HAL CCSD
Résumé : International audience
Modelling techniques for river hydrologic forecasting systems have taken advantage of machine learning methods especially for flood prediction. But current solutions, mostly based on artificial neural networks do not always meet end users requirements on the readability and the understandability of predictive models. In this paper, we present a new version of our original solution based on the concept of aggregate variables in order to predict flash flood events from observed water level and/or rain measurements, particularly in the context of caribbean watersheds in which flash flood are much uncertain. We combine aggregate variables in juries. Juries of aggregate variables are trained and tested using a typical 10-fold cross validation scheme. Best juries are searched through an evolutionary approach that is optimized. Different parameters are set up like aggregation periods and jury sizes to prove the efficiency of the proposed approach compared to classical solutions.
Congress on Evolutionary Computation (CEC) 2013
Cancun, Mexico

hal-00847582
https://hal.archives-ouvertes.fr/hal-00847582
DOI : 10.1109/CEC.2013.6557656

Éditeur(s) : HAL CCSD "Giornale Italiano della Ricerca Educativa - Italian Journal of Educational Research".
Résumé : International audience
The aim of this research is to try to explain beliefs of teachers in favour of student retention, beginning with their psycho-pedagogical beliefs in other areas: learning, intelligence, assessment, principle of justice. A questionnaire was administered to 112 primary teachers of the Belgium French. The responses have been processed through factor analysis to construct and validate metric scales. Then, using regression equations, we have tried to predict beliefs about student retention using other categories of beliefs. Finally, we conducted cluster analyses in order to identify groups of teachers, distinguishable by the nature of their beliefs. Two major results emerge from these analyses. On the one hand, it appears that the awareness of research on the effects of repetition influences teachers’ beliefs about this practice. On the other hand, contrary to our assumptions, the other categories of beliefs do not seem to affect teachers’ beliefs about student retention. These results are discussed in relation to the classical theories postulating that beliefs and social representations are organized into a system.
ISSN: 2038-9744

hal-01532840
https://hal.univ-antilles.fr/hal-01532840

Éditeur(s) : HAL CCSD Association Francophone Internationale de Recherche Scientifique en Education
Résumé : International audience
La question de la structuration des croyances pédagogiques des enseignants en système ou en réseaux est considérée comme essentielle pour comprendre leur fonctionnement professionnel et leurs pratiques. La présente recherche s’inscrit dans cette perspective. Partant de la littérature dans ce domaine, nous avons avancé deux hypothèses principales : 1. Les conceptions des enseignants eu égard à l’apprentissage et à l’intelligence sont liées et affectent leur sentiment d’efficacité pédagogique, en partie par l’intermédiaire des conceptions de l’évaluation ; 2. Tous les enseignants ne partagent pas les mêmes conceptions et croyances, et il est possible de distinguer différents profils selon la façon de les articuler. Pour mettre à l’épreuve nos hypothèses, nous avons conduit une recherche par questionnaire auprès d’environ 200 enseignants du primaire et réalisé une série d’analyses statistiques : modélisations en équations structurales et analyses typologiques. Les résultats confirment partiellement la première hypothèse et mettent en évidence trois profils d’enseignants d’après la configuration des croyances investiguées : « normatif » ; « désengageant » et « actif ».
ISSN: 1647-0117

hal-01532801
https://hal.univ-antilles.fr/hal-01532801