Portail HAL Université de Montpellier Paul-Valéry
Not a member yet
66424 research outputs found
Sort by
Rétrospective réflexive sur l’organisation du travail en interdisciplinarité : un projet relatif au nexus moustique-plastique-eau
No full textInternational audienceOur research project investigates the impact of microplastics on Culex mosquitoes, which are vectors of pathogens. We initially focused on ecology, chemistry, and physics, but then broadened our scope to include the social sciences and humanities. This shift aimed to explore how researchers and stakeholders with various levels of involvement in this project conceived the mosquito-plastic-water nexus. Using semi-structured interviews combined with drawing as a research tool, the study revealed fragmented perspectives of the project, influenced by multiple factors. Some participants felt legitimate only within their specific area of expertise, hesitating to depict aspects beyond their direct involvement. Additionally, limited communication at the project’s outset contributed to a partial view of the latter and a limited understanding of its different parts, steps, and methods. These findings highlight the crucial role of internal communication in fostering a shared understanding among the research team. They also underscore the value of interdisciplinary dialogue, enabling better integration between natural and social sciences to address complex environmental challenges. Finally, they open up the way for interdisciplinary approaches to move towards transdisciplinary approaches.Notre projet de recherche porte sur l’impact des microplastiques sur les moustiques du genre Culex, vecteurs de pathogènes. Nous l’avons d’abord centré sur l’écologie, la chimie et la physique, puis élargi aux sciences humaines et sociales pour mettre en question les représentations que les scientifiques et autres acteurs y étant impliqués avaient du nexus moustique-plastique-eau. Des entretiens semi-directifs, intégrant le dessin comme médium, ont révélé parmi l’équipe des visions fragmentées du projet, influencées notamment par un déficit de communication à son lancement, donnant lieu chez ses membres à un sentiment de légitimité limité à leurs propres domaines d’implication. Cette expérience met ainsi en évidence l’importance d’une communication interne renforcée, tout en ouvrant des réflexions sur l’ouverture des approches interdisciplinaires vers la transdisciplinarité
Annotating Texture and Imitation Patterns in a Corpus of Slow Movements in Corelli’s Trio Sonatas
No full textInternational audienceThe Corelli trio sonatas are emblematic of the mid-Baroque period, characterized by their elegant melodies, intricate counterpoint, and systematic development of thematic material. Extending harmonic annotations from Hentschel et al. (2021), we provide a dataset with annotations of all 38 slow movements of the Sonate da chiesa Op. 1 (1681) and Op. 3 (1689), focusing on this Baroque writing style. The 7,000+ annotations describe the texture (homorhythmy, imitations, suspensions), further detailing thematic and imitation patterns, with other information such as rhythmic density. We provide audio synchronizations to a widely available recording and release the corpus for download as well as through the Dezrann platform
Algebraic Barriers to Halving Algorithmic Information Quantities in Correlated Strings
No full textA preliminary version of the paper was published in proceedings of the 50th International Symposium on Mathematical Foundations of Computer Science (MFCS 2025), https://mfcs2025.mimuw.edu.pl/ ; DOI:10.4230/LIPIcs.MFCS.2025.84, https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.MFCS.2025.84International audienceWe study the possibility of scaling down algorithmic information quantities in tuples of correlated strings. In particular, we address a question raised by Alexander Shen: whether, for any triple of strings (a, b, c), there exists a string z such that each of the values of conditional Kolmogorov complexity C(a|z), C(b|z), C(c|z) is approximately half of the corresponding unconditional Kolmogorov complexity. We provide a negative answer to this question by constructing a triple (a, b, c) for which no such string z exists. Our construction is based on combinatorial properties of incidences in finite projective planes and relies on recent bounds on point-line incidences over prime fields. As an application, we show that this impossibility implies lower bounds on the communication complexity of secret key agreement protocols in certain settings. These results reveal algebraic obstructions to efficient information exchange and highlight a separation in the information-theoretic behavior of projective planes over fields with and without proper subfields
Cooperative learning of Pl@ntNet's Artificial Intelligence algorithm: how does it work and how can we improve it?
No full textInternational audienceDeep learning models for plant species identification rely on large annotated datasets. The PlantNet system enables global data collection by allowing users to upload and annotate plant observations, leading to noisy labels due to diverse user skills. Achieving consensus is crucial for training, but the vast scale of collected data makes traditional label aggregation strategies challenging. Existing methods either retain all observations, resulting in noisy training data or selectively keep those with sufficient votes, discarding valuable information. Additionally, as many species are rarely observed, user expertise can not be evaluated as an inter-user agreement: otherwise, botanical experts would have a lower weight in the AI training step than the average user. Our proposed label aggregation strategy aims to cooperatively train plant identification AI models. This strategy estimates user expertise as a trust score per user based on their ability to identify plant species from crowdsourced data. The trust score is recursively estimated from correctly identified species given the current estimated labels. This interpretable score exploits botanical experts' knowledge and the heterogeneity of users. Subsequently, our strategy removes unreliable observations but retains those with limited trusted annotations, unlike other approaches. We evaluate PlantNet's strategy on a released large subset of the PlantNet database focused on European flora, comprising over 6M observations and 800K users. We demonstrate that estimating users' skills based on the diversity of their expertise enhances labeling performance. Our findings emphasize the synergy of human annotation and data filtering in improving AI performance for a refined dataset. We explore incorporating AI-based votes alongside human input. This can further enhance human-AI interactions to detect unreliable observations
Repenser l’empowerment en santé à l’aune des dispositifs numériques de surveillance de la santé : une approche par le sense-making
No full textInternational audienceThis contribution examines the forms of empowerment mobilized by users of non professional self-tracking (Quantified Self) devices through an analysis of their infor mational and communicational practices. Based on a qualitative study conducted on the MyFitnessPal and Fitbit forums, it explores how individuals interpret the health data produced by these devices and integrate them into their daily routines. The theoretical framework draws on sense-making theory and the sociology of uses to investigate processes of appropriation, contestation, and negotiation of digital norms. Three forms of empowerment are identified: cognitive (the ability to interpret and adapt data), critical (distancing oneself from algorithmic authority), and collective (peer support and learning). Far from a purely individualistic model, these forms of empowerment are situated, relational, and sometimes conflictual. The study thus highlights the complexity of health-related connected device uses, emphasizing their role as sociotechnical mediation devices where relationships to the body, information, and autonomy are continuously reconfigured.Cette contribution examine les formes d’ empowerment mobilisées par les usagers de dispositifs non professionnels de quantification de soi (quantified self) à travers l’analyse de leurs pratiques informationnelles et communicationnelles. En s’appuyant sur une enquête qualitative menée sur les forums MyFitnesspal et Fitbit , il explore la manière dont les individus interprètent les données de santé produites par ces dispositifs et les intègrent à leurs routines. L’approche théorique mobilise les apports du sense-making et de la sociologie des usages pour interroger les processus d’appropriation, de contestation et de négociation des normes numériques. Trois formes d’empowerment sont mises en évidence : cognitif (capacité à interpréter et à adapter les données), critique (mise à distance de l’autorité algorithmique) et collectif (soutien et apprentissage par les pairs). Loin d’un modèle purement individualiste, ces formes d’empowerment sont situées, relationnelles et parfois conflictuelles. L’étude met ainsi en lumière la complexité des usages des objets connectés de santé, en soulignant leur rôle de dispositifs de médiation sociotechnique où se rejouent les rapports au corps, à l’information et à l’autonomie
La conscientisation des imaginaires comme processus d’appropriation du français chez des étudiants brésiliens. Contribution à une didactique des langues critique
No full textInternational audienceFrom a Freirean perspective of dialogic and critical education, this article highlights the central role of raising awareness about the social imaginary (Castoriadis, 1975) in the process of Brazilian students appropriating the French language. The two cases of conscientization presented here are based on a critical discourse analysis of Brazilian students’ narratives collected during two self-reflective activities: a testimony about their relationship with a particular language followed by a discussion, and an excerpt from their biographical language interview. Through these elicitation and dialogue activities—exploring their relationships with themselves, the Other, languages, knowledge, and learning—the students reveal to their language teacher the instituted imaginary that oppresses them. This process enables the teacher to recognize her own historicity and the vulnerability of her students. This emancipatory classroom practice ultimately allows both the students and the teacher to acknowledge themselves as historico-social beings, co-constructors of knowledge, and reciprocal educators. As such, it aligns with a critical language education.Dans une perspective freirienne d’éducation dialogique et critique, cet article met en évidence le rôle central de la conscientisation de l’imaginaire social (Castoriadis, 1975) dans la démarche d’appropriation du français d’étudiants brésiliens.Les deux cas de conscientisation présentés dans cet article reposent sur l’analyse critique de discours d’étudiantes brésiliennes saisis dans deux activités ayant pour but de faire un récit de soi : un témoignage sur le rapport à une langue en particulier suivi d’un échange, puis un extrait d’interview biographique langagière. À travers ces activités d’élicitation et de dialogue sur leurs rapports à elles, à l’Autre, aux langues, au savoir et à l’apprentissage, les étudiantes révèlent à leur enseignante l’imaginaire institué qui les opprime, permettant à cette dernière de prendre conscience de sa propre historicité, et de leur vulnérabilité. Cette pratique de classe émancipatrice permet in fine aux étudiantes comme à l’enseignante de se reconnaitre comme des êtres historico-sociaux, co-constructrices des savoirs, et éducatrices réciproques. Elle s’inscrit ce faisant dans une didactique des langues critique
: Urbanization and professional retraining of market gardeners on the outskirts of Dakar (Senegal)
No full textInternational audienc
Les identifiants persistants dans le schéma de métadonnées Datacite : recommandations du consortium français
No full textThis document was produced by a working group of the French DataCite consortium led by INIST-CNRS. It complements those produced by the consortium's other working groups (Metadata, Scope, Economic Model, Media, Awareness and Training, and Communication).The aim is to provide some guidelines on the choice of persistent identifiers that can be declared when creating DOIs in DataCite and associated metadata. It is intended for publication managers, particularly repository managers.These recommendations are only indicative, as the landscape of persistent identifiers for research is constantly evolving. They are aligned with those of the MESRE (Ministry of Higher Education, Research, and Space) regarding persistent identifiers.Ce document a été réalisé dans le cadre d’un groupe de travail du consortium français de DataCite porté par l’INIST-CNRS. Il est complémentaire à ceux réalisés par les autres groupes de travail du consortium (Métadonnées, Périmètre, Modèle économique, Supports, sensibilisation et formations et Communication).L’objectif est de proposer quelques lignes directrices sur le choix des identifiants persistants pouvant être déclarés dans le cadre de la création de DOI dans DataCite et des métadonnées associées Il s’adresse aux responsables de publication, et notamment les responsables d’entrepôts.Ces recommandations ne peuvent être qu’indicatives, le paysage des identifiants persistants pour la recherche étant en évolution constante. Elles s’alignent sur celles du MESRE (Ministère de l'Enseignement supérieur, de la Recherche et de l'Espace) au sujet des identifiants persistants