Éditeur(s) : HAL CCSD
Résumé : International audience
In this paper, we describe a blood alcohol content estimation prototype based on a comportment analysis performed by artificial neural networks. We asked to subjects that had drunk alcohol to play a video-game after having measured their blood alcohol content with a breathalyser. A racing game was modified so that it could provide various data related to the use of the controls by the player. Using the collected data, we trained our neural network in order to be able to determine whether or not the subject had exceeded a blood alcohol content threshold. We also succeeded in estimating this blood alcohol content with a mean error of 0.1g/l. We could perform those estimations independently of the track played among the two ones used. It was also performed in "real time", e.g., using only the data collected within the last 10 seconds of playing.
Proceedings of the 2011 International Conference on Advances in Computer-Human Interactions
Gosier, Guadeloupe

Droits : info:eu-repo/semantics/OpenAccess
hal-00849971
https://hal.archives-ouvertes.fr/hal-00849971
https://hal.archives-ouvertes.fr/hal-00849971/document
https://hal.archives-ouvertes.fr/hal-00849971/file/article.pdf

Éditeur(s) : HAL CCSD Ecological Society of America
Résumé : Reliable assessment of fish origin is of critical importance for exploited species, since nursery areas must be identified and protected to maintain recruitment to the adult stock. During the last two decades, otolith chemical signatures (or "fingerprints") have been increasingly used as tools to discriminate between coastal habitats. However, correct assessment of fish origin from otolith fingerprints depends on various environmental and methodological parameters, including the choice of the statistical method used to assign fish to unknown origin. Among the available methods of classification, Linear Discriminant Analysis (LDA) is the most frequently used, although it assumes data are multivariate normal with homogeneous within-group dispersions, conditions that are not always met by otolith chemical data, even after transformation. Other less constrained classification methods are available, but there is a current lack of comparative analysis in applications to otolith microchemistry. Here, we assessed stock identification accuracy for four classification methods (LDA, Quadratic Discriminant Analysis [QDA], Random Forests [RF], and Artificial Neural Networks [ANN]), through the use of three distinct data sets. In each case, all possible combinations of chemical elements were examined to identify the elements to be used for optimal accuracy in fish assignment to their actual origin. Our study shows that accuracy differs according to the model and the number of elements considered. Best combinations did not include all the elements measured, and it was not possible to define an ad hoc multielement combination for accurate site discrimination. Among all the models tested, RF and ANN performed best, especially for complex data sets (e. g., with numerous fish species and/or chemical elements involved). However, for these data, RF was less time-consuming and more interpretable than ANN, and far more efficient and less demanding in terms of assumptions than LDA or QDA. Therefore, when LDA and QDA assumptions cannot be reached, the use of machine learning methods, such as RF, should be preferred for stock assessment and nursery identification based on otolith microchemistry, especially when data set include multispecific otolith signatures and/or many chemical elements.
ISSN: 1051-0761

hal-00644549
https://hal.archives-ouvertes.fr/hal-00644549
DOI : 10.1890/09-1887.1

Éditeur(s) : HAL CCSD
Résumé : International audience
This paper demonstrates that a human being using an interface can be efficiently evaluated - in real time - by embedding basic measurements in the interface and using a suitable trained artificial neural network. The approach is introduced through video games but is suitable for any machine capable of valuable measurements on user actions. Of course, the quality of the "diagnostic" depends of the learnability of the task and of the size and quality of the learning base. Typical applications include the detection of fatigue, stress, emotions, the influence of a drug or of medical treatments ; screening a deficit or adequateness to a task, etc. Two successful prototypes are presented, one to predict the mental age of children through a set of simple basic games, and the other to detect if a subject is right-handed of left-handed through a racing car simulation.
Proceeding of the Third International Conferences on Advances in Computer-Human Interactions
St. Maarten, Netherlands Antilles

hal-00601476
https://hal.archives-ouvertes.fr/hal-00601476

Résumé : Les fourmis sont des organismes cibles appropriés pour les études environnementales. Au cours de notre étude, nous nous sommes focalisés sur les fourmis de la litière via l'application de deux méthodes de récolte complémentaires selon le protocole all (ants of leaf litter). Dans les forêts naturelles de Guyane française, l'hétérogénéité environnementale et les perturbations structurent les communautés de fourmis de la litière en influençant la richesse, la diversité, l'abondance et la densité en espèces ainsi que les compositions taxonomique et fonctionnelle. Chaque type de formation végétale possède une communauté spécifique. La fragmentation et la conversion des forêts en plantations ont entrainé une altération plus ou moins profonde des communautés; cette variabilité est fonction du type d’agriculture et des espèces d'essences cultivées. Bien que l'altération des communautés se soit révélée maximale dans la plupart des monocultures, les plantations de cacaoyers ont un potentiel de conservation réel. Enfin, dans un contexte ou il est urgent de simplifier l'intégration des arthropodes dans les études de conservation, de contrôle et de suivi de la santé des écosystèmes terrestres, la méthode des réseaux de neurones est apparue comme un outil puissant pour mettre en évidence et analyser les patrons des communautés de fourmis.
Ants are reliable and relevant target organisms for environmental surveys. In our study, we focused on litter-dwelling ants through the use of two complementary sampling methods that were implemented according to the ants of leaf litter (all) protocol. In pristine forests of french guiana, leaf-litter ant communities are structured by environmental heterogeneity and natural perturbations - which influence species richness, diversity, abundance and density, as well as taxonomic and functional composition. This results in habitat-specific communities for each vegetal formation. Forest fragmentation and conversion into monocultures have led to a more or less deep alteration of the ant communities; this variability depends on the type of agricultural system and cultivated tree species. Although the intensity of community alteration peaked in tree monocultures, cocoa plantation exhibited a real potential for native species conservation. Finally, in the current context where simplifying the integration of arthropods into conservation programs as well as into surveys designed to monitor and manage the environment is critical, the use of neural networks appears to be a powerful tool for reliably highlighting and analyzing ant communities patterns.

http://www.theses.fr/2011AGUY0475/document