HAL-uB
Not a member yet
73989 research outputs found
Sort by
Analyser les résultats des appels à projets des fonds pédagogiques universitaires par le biais d’une cartographie évaluative : Application au NCU RITM-BFC en France
No full textInternational audienceCette communication s’intéresse aux appels à projets (AAP) soutenant l’innovation et le développement pédagogiques dans les universités via les fonds pédagogiques, dispositif clé complétant d’autres formes de soutien aux enseignants (microprogrammes, centres de soutien ou congés sabbatiques pédagogiques). Bien souvent, l’évaluation de ces AAP sont réalisés par le biais d’un compte-rendu administratif dans une logique de reddition de comptes aux instances, en comportant des indicateurs d’activités (nombre d’actions, nombre de personnes…), ainsi que des indicateurs financiers. Dans certains cas, d’autres informations peuvent être données concernant le fonctionnement, les effets des actions réalisées etc., sous l’impulsion des exigences des financeurs. Cette forme d’évaluation administrative synthétique, plus sommative que formative, peut dépasser cette logique initiale pour nourrir la réflexion de la gouvernance, avant même la mise en place des actions sur le terrain et leur clôture. C’est précisément l’objet de cette communication, qui propose une approche cartographique qualitative nourrie par les avancées de la recherche en éducation. Mise en œuvre dans le cadre de l’évaluation du dernier AAP de RITM-BFC (24 projets financés en 2025 pour soutenir la réussite étudiante), cette démarche repose sur l’analyse des informations contenues dans les dossiers de candidature et sur les diagnostics établis avec les porteurs de projets et/ou l’équipe opérationnelle du fonds, à partir d’une grille descriptive des dispositifs pédagogiques. Une cartographie générale permet ainsi d’éclairer la réflexion sur la logique « bottom-up » adoptée par ce fond pédagogique et à alimenter la réflexion stratégique sur son pilotage. Cette méthodologie est ainsi testée et discutée dans le cadre de RITM-BFC. En lien avec l’axe 4 du colloque, cette communication propose d’enrichir les pratiques des évaluateurs avec une méthodologie peu usitée en se plaçant à l’échelle d’un programme ayant des projets variés
"How many countries are in this country?": Representing Britishness in Apple TV+'s Ted Lasso (2020-2023)
No full textInternational audienceTed Lasso is a series rooted in Anglophone culture clash: Jason Sudeikis originated the titular character as the feckless protagonist of a series of ads, where an American football coach finds himself the new coach for an English football/soccer club. Ted Lasso’s focus on its fish-out-of-water hero and his experience of Britishness allowed Apple TV+ to broaden its focus to a more international audience, a sort of gateway to more varied affiliations, highlighting not just the interactions between the American and British characters, but the array of international players (among them Wales, the source of Ted’s question that makes up our title). In so doing, the series arguably echoes American audiences’ own experience of international fictions. However, it also insists on Britain as a cosmopolitan society, contrasting the cozy Englishness of heritage site Paved Court, Richmond with a population that reflects the varied nature of British identity, and where the focus is on cultural inclusion. It is this tension (between American and British, between English and British, between heritage and multiculturalism) that this article would seek to examine
Symmetries of -foams
No full textv1: 34 pages, many figures. Comments welcome! v2: 35 pages, exposition improved following referee's commentsWe give an action of a Lie subalgebra of the Witt algebra on foams. This action is compatible with the -foam evaluation formula. In particular, this endows states spaces associated with -webs with an -action. When working in positive characteristic, this can be used to define a -DG structure on these state spaces
Sleep Stages Classification with Multi-modal Signals using Deep Learning
No full textInternational audienceAccurate sleep stage classification is essential for diagnosing and managing sleep disorders. This study presents a novel deep learning architecture for automatically classifying sleep stages. Rigorous evaluation on the SOMEG dataset yielded a notable 82.05% overall accuracy, significantly outperforming traditional machine learning approaches. Comprehensive analysis of training, validation, and test accuracy metrics demonstrates the model's robust generalization across sleep stages. While confusion matrix analysis revealed potential refinements, particularly in differentiating EEG patterns with higher individual variability, these findings underscore the substantial potential of our deep learning framework for precise and automated sleep stage classification.</div
CALPHAD-aided synthesis and characterization of an Al–Co–Cr–Fe–Ni–W high-entropy alloy prepared by arc melting and spark plasma sintering
No full textInternational audienceIn the present work, the novel Al0.15CoCrFeNiW0.15 High-Entropy Alloy (HEA) has been designed by CALPHAD (CALculation of PHAse Diagrams) computations with the in-house built Genova High-Entropy Alloys (GHEA) database, aiming to a mostly monophasic face-centered cubic (FCC) alloy strengthened by the precipitation of secondary μ phase. To explore different preparation routes, alloy samples have been synthesized by both arcmelting (AM) and spark plasma sintering (SPS). Samples were characterized by low optical microscopy (LOM), scanning and transmission electron microscopy (SEM and TEM), X-ray diffraction (XRD), and microhardness measurements. Long-term annealing at 1100 °C has been performed, followed by quenching or furnace cooling. AM as-cast sample showed a monophasic FCC microstructure, characterized by large grains. Precipitation of μ phase was observed in the equilibrated and quenched sample, in good agreement with the thermodynamic calculations. On the other hand, SPS samples resulted in a finer microstructure, characterized by the presence of small particles of Al2O3 and μ phase, already present before annealing. Contrary to the thermodynamic predictions, after equilibration and quenching, the dissolution of the μ phase was observed due to the Gibbs-Thomson effect, which enhanced W solubility in the FCC solid solution. Annealing of the SPSed alloy followed by furnace cooling, however, allowed the precipitation of μ, thanks to the slower cooling rate. Overall, this study highlighted CALPHAD’s utility for composition selection in complex multicomponent systems anddemonstrated how AM and SPS lead to significantly different microstructures and properties, with grain size playing a key role in determining the alloy performances
Mapping applications of laser-laser isotopic measurement in carbonates
No full textInternational audienceSedimentary geochemistry is very often associated with the measurement of isotopic composition of carbon and oxygen from carbonates. The usual technique combining acid digestion and mass spectrometry analysis is slow, costly and non-ideal for spatially resolved analyses. When carbonates are processed using laser calcination and the gas produced during calcination is analyzed by infrared spectrometry, the time required for isotopic analysis is reduced to around 15 min to analyze 30 mg of carbonate in situ. Although the time saved is significant, it is hardly reasonable to carry out a high-resolution isotopic mapping of large samples. A fully resolved isotopic mapping, for example, of a sample with 25 cm2 surface area at resolution of a tenth of a millimeter, would require a continuous measurement carried out for a month. The aim of this study is, therefore, to explore possible strategies for constructing an isotopic map with a minimum number of analyses. Two approaches are pro-posed: (i) a mathematical approach that seeks to establish a correlation between the position of the sample and the carbon or oxy-gen isotopes, and (ii) an approach that looks for a correlation between the color (spectral characteristics) of the sample surface subdomains and their isotopic compositions. The choice of the second approach stems from the assumption that color contains a priori information about geological or geochemical processes. Several algorithms were developed and tested, notably using artificial intelligence tools. To testify the isotopic maps produced by these algorithms, posteriori isotopic measurements are taken and compared with the predictions from computed isotopic maps
Pratiques ordinaires et reconnues de l’engagement étudiant pour l’environnement : articulation aux formations et logiques d’usage dans l’expérience étudiante
No full textNational audienc
How deep is your love? Explore user attachment to digital sharing platforms
No full textInternational audienceWhile sharing platforms have shown their continued rise and growing importance in recent years, they now operate in competitive market conditions and a changing environment: scam issues (Vinted platform), legal evolution (Airbnb issue). Today, it is becoming essential for digital sharing platforms to develop strong and durable relationships with users. In this context, platform attachment is the key to success (Li et al., 2022). This study explores a concept widely used in marketing literature, brand attachment, within the context of sharing platforms. A qualitative study was conducted among users of the HomeExchange platform to identify andbetter understand the platform attachment construct. Our exploratory study reveals five user profiles and suggests that while user experiences impact satisfaction, which leads to platform attachment, user compatibility remains the key factor. This study contributes to the marketing literature by investigating the role of user attachment to sharing platforms and provides a holistic understanding of how platform attachment develops
L'empreinte du locuteur : subjectivation progressive du discours présidentiel de Donald Trump
No full textInternational audienceNous examinons l’évolution sémantique des discours présidentiels de Donald Trump au cours de son premier mandat (2017–2021), sur la base de méthodes quantitatives issues de la linguistique de corpus et du traitement automatique du langage. Notre corpus comprend l’ensemble des interventions officielles transcrites, segmentées en unités de discours homogènes, puis analysées par word embedding. Deux échelles d’évolution sont considérées : une dynamique externe, mesurée à l’échelle du mandat, et une dynamique interne, observée au sein même de chaque discours. L’analyse externe ne révèle pas de déplacement sémantique très marqué dans le temps, ce qui suggère une stabilité relative du registre présidentiel. En revanche, une évolution interne claire se manifeste : les discours de Trump débutent dans un registre institutionnel et impersonnel, puis évoluent vers une formulation plus subjective, marquée par l’usage croissant de la première personne et une baisse de densité lexicale. Ce schéma indique une structuration spécifique du discours présidentiel trumpien, caractérisée par une transition progressive du rôle représentatif vers l’expression personnelle
Hivers plus doux, printemps changeants : la dynamique de résistance au froid prédit le débourrement et les risques liés aux basses températures à l’échelle intercontinentale
No full textInternational audienceTemperate woody perennial plants form buds during late summer that contain leaves and flowers that emerge in the following growth season. To survive winter, dormant buds must attain cold hardiness, and timely lose it in spring to break bud while avoiding damage from low temperatures and late frosts. Here, we use an untrained process-based model to predict bud cold hardiness of three grapevine varieties ( V. vinifera 'Cabernet-Sauvignon' and 'Riesling', and V. hybrid 'Concord') from historical temperature records of eight different locations in North America and Europe (n = 329). Based on those predictions, and thresholds of cold hardiness at budbreak from literature, timing of budbreak was extracted. Despite being untrained to the data, the RMSE of budbreak predictions was 7.3 days (Bias=−0.83). Based on cold hardiness estimations and air temperature records, low temperature damage was quantified and validated through newspapers and extension records. In years × location where damage was predicted, corrections to budbreak based on delays expected resulted in improvements of predictions (RMSE=7.2d, Bias=0.58). Predictions of instances of freeze damage risk demonstrate genotypic adaptation to different environments. At the species level, increasing or decreasing trends in freeze damage risk are predicted, depending on the range of mean dormant season temperature (MDST; 1 Nov - 30 Apr) present in each location. Sensitivity analysis of predicted time to budbreak based on MDST shows a general advancement of phenology at −5.8d/°C. However, in much warmer locations, delays can be expected as temperatures continue to increase (+1.9d/°C for MDST>10°C). Through cold hardiness dynamics, the estimation of chilling accumulation appears as an important source of error for predictions of spring phenology across environments. Cold dynamics represents an advancement in phenological modeling that provides information for the entirety of the dormant season, as well as budbreak