Teses / dissertações sobre o tema "Intelligence Artificielle en Éducation"
Crie uma referência precisa em APA, MLA, Chicago, Harvard, e outros estilos
Veja os 50 melhores trabalhos (teses / dissertações) para estudos sobre o assunto "Intelligence Artificielle en Éducation".
Ao lado de cada fonte na lista de referências, há um botão "Adicionar à bibliografia". Clique e geraremos automaticamente a citação bibliográfica do trabalho escolhido no estilo de citação de que você precisa: APA, MLA, Harvard, Chicago, Vancouver, etc.
Você também pode baixar o texto completo da publicação científica em formato .pdf e ler o resumo do trabalho online se estiver presente nos metadados.
Veja as teses / dissertações das mais diversas áreas científicas e compile uma bibliografia correta.
Querrec, Ronan. "Les Systèmes Multi-Agents pour les Environnements Virtuels de Formation : Application à la sécurité civile". Brest, 2003. http://www.theses.fr/2002BRES2037.
Texto completo da fonteThis study concerns virtual environments for training in operationnal conditions. The principal developed idea is that these environments are heterogeneous and open multiagent systems. The MASCARET model is proposed to organize the interactions between the agents and to give them reactives, cognitives ans socials abilities to simulate the physical and social environment. The physical environment represent, in a realistic way, the phenomena that the learners and the teachers have to take into account. The social environment is simulated by agents executing collaborative and adaptative tasks. They realize, in team, the procedures that they have to adapt to the environment. The users participate to the training environment throught their avatar. To validate our model, the SecuRevi application for fire-fighters training is developed
Delorme, Fabien. "Evaluation et modélisation automatiques des connaissances des apprenants à l'aide de cartes conceptuelles". Rouen, INSA, 2005. http://www.theses.fr/2005ISAM0012.
Texto completo da fonteDimitracopoulou, Angélique. "Le tutorat dans les systèmes informatisés d'apprentissage : étude de la conception et réalisation d'un tutoriel d'aide à la représentation physique des situations étudiées par la mécanique". Paris 7, 1995. http://www.theses.fr/1995PA070089.
Texto completo da fonteThe modelisation of tutorials process, a crucial question in the field of computational didactics, is the main question of our research in design and realization of the prototype arpia : it is a tuturial of aid for the physics representation of situations in mechanics ; it is interactive and adapted to the learner. Our target is to help the students (between 17-19 years old) who face difficulties in developping a physics representation of situations (diagrams of forces and movements). The theoretical framework of the design of this tutorial is based on a precisedidactic analysis : epistemological and cognitive analysis of content, analysis of students' difficulties, clarification of learning hypothesis. The interface of communication has been designed in order to permit to the student to construct his representations directly (for instance by drowing the vectors). The process of cognitif diagnosis identifies and interprets the elements of student's representations. The prototype arpia has actually eight types of action : for instance, advice for a use of atechnic, procedure of correction of errors, process of explanation
Gaguet, Laurent. "Attitudes mentales et planification en intelligence artificielle : modélisation d'un agent rationnel dans un environnement multi-agents". Clermont-Ferrand 2, 2000. http://www.theses.fr/2000CLF20023.
Texto completo da fonteVinurel, Jean-Jacques. "Une application de l'intelligence artificielle à l'enseignement assisté par ordinateur". Paris 6, 1986. http://www.theses.fr/1986PA066379.
Texto completo da fonteSnell, Pierre. "Aide intelligente pour la surveillance d'examens". Master's thesis, Université Laval, 2020. http://hdl.handle.net/20.500.11794/66431.
Texto completo da fonteNowadays, more and more courses are available online. On the other hand, they remain marginal and few of them have given valid and recognized certifications. Yet, those lessons have incredible content, and are an amazing resource. Furthermore, more and more, efforts are made toward numerising learning. The goal of this thesis is to improve online exam monitoring. Indeed, one of the strong points of the internet is anonymity, and attesting the success and skills of someone quickly becomes non-trivial. Moreover, the only commonly accessible tools available for surveillance are the camera, and the user’s keyboard / mouse which gives very sparse information. To do so, we used the most up-to-date tracking, recognition, generation and analysis techniques to introduce automation and «intelligence» in the surveillance. Note that even for a human this task is difficult and that to decide if a behavior is abnormal is not deterministic. The goal is therefore to provide help and support to bring large-scale examination monitoring to a safer level. The method described is based on the advances in artificial intelligence, particularly generative adversials network (GANs), reinforcement learning (RL) convolution neural networks (CNN) and recurrent neural networks (RNN).
Darwesh, Aso. "Diagnostic cognitif en EIAH : le système PépiMep". Paris 6, 2010. http://www.theses.fr/2010PA066397.
Texto completo da fonteCourtois, Joël. "Siam : un système de diagnostic qui s'adapte aisément à de nouveaux domaines et qui enseigne sa méthode". Paris 6, 1990. http://www.theses.fr/1990PA066461.
Texto completo da fonteLabat, Jean-Marc. "Quiz : une contribution à l'amélioration des capacités pédagogiques des tuteurs intelligents : sous-titre : une étude réalisée sur l'enseignement des enchères au bridge". Paris 6, 1990. http://www.theses.fr/1990PA066566.
Texto completo da fonteGrumbach, Alain. "Contribution à l'étude de modèles d'apprentissage en interaction avec un environnement inspirés du comportement humain". Paris 11, 1987. http://www.theses.fr/1987PA112244.
Texto completo da fonteThe aim of this research is to design and write programs which provide machines with learning by doing capabilities, in a problem solving situation. The approach is inspired by human behavior study : subject observations, model building, drawing out of fundamental notions, design of corresponding software (language programs). The fundamental notions set out in this work concern : knowledge types and relationship, which gave rise to the knowledge Lattice structure that relates general and specific knowledge fields to one another ; learning behavior trigerring : on event ; acquired informations : links between them,. . . The language built up from these notions is twofold : Multilog, implementation of the Knowledge Lattice within Logic Programing through clause "worlds" ; Evenlog, implementation of the event notion. Different human learning simulation programs have been developed, ranging from a specific model to a very general one whose unreached objective was to be able to learn non specific knowledge fields. Lastly, this work, as part of Cognitive Science Research, proposes a short Cogitation on Knowledge : the above mentionned Knowledge Lattice ; different points of view : behaviorist, cognitive,. . . , links between them ; (phylo) genesis
Buche, Cédric. "Un système tutoriel intelligent et adaptatif pour l'apprentissage de compétences en environnement virtuel de formation". Phd thesis, Université de Bretagne occidentale - Brest, 2005. http://tel.archives-ouvertes.fr/tel-00011223.
Texto completo da fonteDans ce contexte, nous soutenons la thèse qu'il est possible d'intégrer un système tutoriel intelligent (ITS) générique et adaptatif dans un environnement virtuel afin de fournir une aide pédagogique à l'apprenant et une assistance pédagogique au formateur.
Cette thèse débute par une étude montrant l'intérêt mutuel de la réalité virtuelle et des ITS pour l'apprentissage de compétences, et identifie les difficultés de leur intégration. Plus précisément, elle souligne la nécessité d'une représentation abstraite, indépendante de l'exercice à réaliser, manipulable pour la prise de décision pédagogique et liée à la représentation d'un univers 3D.
Notre proposition est un système multi-agents permettant d'analyser l'action réalisée par l'apprenant par le biais d'un environnement virtuel informé. Le système dégage un ensemble d'informations, appelé situation pédagogique, considéré pertinent pour la prise de décision pédagogique. Notre étude se focalise alors sur un agent pédagogique qui propose des assistances au formateur en utilisant la situation pédagogique. L'abstraction utilisée permet des assistances concrètes liées au domaine, à l'exercice et à l'environnement virtuel. Le modèle comportemental de l'agent pédagogique se base sur un système de classeurs hiérarchique. Grâce à ce modèle, l'agent s'adapte au couple apprenant-formateur en modifiant son comportement pédagogique par le biais d'un mécanisme d'apprentissage artificiel, basé sur un renforcement fourni par le formateur.
Ces travaux sont appliqués dans le cadre du projet GASPAR (Gestion de l'Activité aviation et des Sinistres sur Porte-avions par la Réalité virtuelle). L'application simule l'activité aviation sur un porte-avions.
Zaarour, Iyad. "Contributions à la découverte des stratégies d'écriture d'élèves de scolarité primaire". Rouen, 2004. http://www.theses.fr/2004ROUES001.
Texto completo da fonteThe aim of this study is to bring a contribution to the realization of the evolution follow-up in writing among typical pupils in primary education. For this purpose, we have developed a software for the acquisition of handwritten tracings and the automatic extraction of features from these tracings. Distributed on three periods of about six months each, the acquisitions have therefore been achieved three times for the same pupils in the same experimental conditions, these tracings being acquired online by means of a digitizer. An unsupervised classification is first applied on a set of dynamic features chosen by an expert in the field of child's development psychology; strong forms are thus selected as steady clusters from the obtained partitions. With this unsupervised approach, we have thus discovered three strategies: a first one which is performant in control and global planning, a second one labeled local in control and planning, and a third one which is an unstable intermediary strategy. Next we modeled our problem by means of a probabilistic graphical model (bayesian network) in which the writing strategy is represented by a hidden variable. We build a global hierarchical model in order to link local and global strategies and model the probabilistic dependance between variables and strategies. Our hierarchical model, learnt with real data, enables us to discover two global strategies that correspond to normo-writer pupils and more advanced normo-writers. These two strategies are consistent: the distribution of typical pupils by school level is constant over time, and the probability of transition between (or within) these strategies is also constant over time
Tran, Tri Duc. "Conception et développement d'un assistant intelligent pour un accompagnement conatif des élèves en difficultés". Paris 6, 2010. http://www.theses.fr/2010PA066099.
Texto completo da fonteMontagne, Fabien. "Une architecture logicielle pour aider un agent apprenant par renforcement". Littoral, 2008. http://www.theses.fr/2008DUNK0198.
Texto completo da fonteThis thesis deals with reinforcement learning. One of the main advantage of this learning is to not require to know explicitely the expected behavior. During its learning, the agent percieves states, gets a set of rewards and selects actions to carry out. The agent fits its behavior by optimizing the amount of rewards. Nevertheless, the computing time required quickly becomes prohibitive. This is mainly due to the agent’s need of exploring its environment. The approach considered here consists in using external knowledge to “guide” the agent during its exploration. This knowledge constitutes an help which can, for example, be expressed by trajectories that set up a knowledge database. These trajectories are used to limit the exploration of the environment while allowing the agent to build a good quality behavior. Helping an agent does neither involve knowing the actions choose in all states, nor having the same perceptions as the agent. The critic-critic architecture was devised to fulfill to this problematic. It combines a standard reinforcement learning algorithm with an help given through potentials. The potentials assiociate a value to each transition of the trajectories. The value function estimation by the agent and the potential of the help are combined during the training. Fitting this combine dynamically makes it possible to throw assistance into question while guaranteing an optimal or almost optimal policy quickly. It is formally proved that the proposed algorithm converges under certain conditions. Moreover, empirical work show that the agent is able to benefit from an help without these conditions
Boufflet, Jean-Paul. "Emplois du temps dans un environnement fortement contraint : exemple de l'U.T.C". Compiègne, 1992. http://www.theses.fr/1992COMP454S.
Texto completo da fonteBoufflet, Jean-Paul. "Emplois du temps dans un environnement fortement contraint : exemple de l'U.T.C". Compiègne, 1992. http://www.theses.fr/1992COMPD454.
Texto completo da fonteMacrelle-Rosselle, Marilyne. "Conception d'un atelier d'expérimentation de logiciels éducatifs : application en géométrie". Phd thesis, Université Henri Poincaré - Nancy I, 2001. http://tel.archives-ouvertes.fr/edutice-00000263.
Texto completo da fonteNarassiguin, Anil. "Apprentissage Ensembliste, Étude comparative et Améliorations via Sélection Dynamique". Thesis, Lyon, 2018. http://www.theses.fr/2018LYSE1075/document.
Texto completo da fonteEnsemble methods has been a very popular research topic during the last decade. Their success arises largely from the fact that they offer an appealing solution to several interesting learning problems, such as improving prediction accuracy, feature selection, metric learning, scaling inductive algorithms to large databases, learning from multiple physically distributed data sets, learning from concept-drifting data streams etc. In this thesis, we first present an extensive empirical comparison between nineteen prototypical supervised ensemble learning algorithms, that have been proposed in the literature, on various benchmark data sets. We not only compare their performance in terms of standard performance metrics (Accuracy, AUC, RMS) but we also analyze their kappa-error diagrams, calibration and bias-variance properties. We then address the problem of improving the performances of ensemble learning approaches with dynamic ensemble selection (DES). Dynamic pruning is the problem of finding given an input x, a subset of models among the ensemble that achieves the best possible prediction accuracy. The idea behind DES approaches is that different models have different areas of expertise in the instance space. Most methods proposed for this purpose estimate the individual relevance of the base classifiers within a local region of competence usually given by the nearest neighbours in the euclidean space. We propose and discuss two novel DES approaches. The first, called ST-DES, is designed for decision tree based ensemble models. This method prunes the trees using an internal supervised tree-based metric; it is motivated by the fact that in high dimensional data sets, usual metrics like euclidean distance suffer from the curse of dimensionality. The second approach, called PCC-DES, formulates the DES problem as a multi-label learning task with a specific loss function. Labels correspond to the base classifiers and multi-label training examples are formed based on the ability of each classifier to correctly classify each original training example. This allows us to take advantage of recent advances in the area of multi-label learning. PCC-DES works on homogeneous and heterogeneous ensembles as well. Its advantage is to explicitly capture the dependencies between the classifiers predictions. These algorithms are tested on a variety of benchmark data sets and the results demonstrate their effectiveness against competitive state-of-the-art alternatives
Ali, Houssein Souleiman. "Assistance à l’apprentissage de l’algorithmique : méthode et outil pour l’évaluation et la rétroaction". Thesis, Lille, 2019. http://www.theses.fr/2019LIL1I064.
Texto completo da fonteMastering the basic concepts of programming and the ability to carry out simple programs are the essential objectives of introductory computer science courses. The teaching and learning of programming is considered complex, which explains the high drop-out rate of these programs, widely documented in the literature. Many works provide learning environments for programming that assist learners in mastering the syntax and semantics of programming languages. However, among the causes of failure, the literature identifies a lack of ability to decompose and formalize a problem in the form of an algorithm.In the context of this thesis, we are interested in the analysis and algorithmic solution phase. The purpose of this phase is to teach the learner a problem solving and formalization process. In this phase the learner describes or structures his algorithmic solution using a formal notation (pseudo-code) independent of any programming language.The literature review indicates that improving problem solving skills requires practice. As part of programming for novices, it is necessary to do many exercises (problem solving) with increasing difficulty levels. It is also necessary to provide feedback related to the mistakes made by the learners.For this purpose we propose AlgoInit, a Web environment for algorithmic learning. Based on the study of the literature, we defined:- A modeling based on Bloom's taxonomy to define the cognitive level of the proposed exercises.- An evaluation approach based on the comparison of a learning solution to a model solution. To compare the solutions (learner and models) described in pseudo-code, we go through a step of transforming solutions into labeled trees built from a rule base- Rules for providing feedback and progressive exercises based on the outcome of the learner's solution assessment.To evaluate the potential of our prototype, we have conducted two experiments at the University of Djibouti. The first experiment was devoted to evaluating the ability of our prototype to recognize the different algorithmic solutions. As for the second one, it was devoted to the evaluation of the educational interest of AlgoInit. These experiments have shown convincing results on the ability of AlgoInit to classify solutions (correct and incorrect) and to provide useful feedback. The results also indicate that our system has a significant influence on students’ problem solving ability
Eleuterio, Marco Antonio. "AMANDA : a computational method for mediating asynchronous group discussions". Compiègne, 2002. http://www.theses.fr/2002COMP1400.
Texto completo da fonteThis thesis is about computer-mediated communication. It describes AMANDA, a computational method for mediating asynchronous group discussions among distant leamers. The proposed method is intended to coordinate collective discussions and improve group communication with negligible or no human effort. The method consists of launching a set of issues for collective debate and involving the participants in successive discussion cycles. At each cycle, the answers and arguments provided by the participants are intentionally redistributed among the group. Throughout the discussion, specifie mechanisms' search for potential interactions that might improve the debate and propose new interactions among the group. Ln addition to the intelligent mediation of group discussions, AMANDA supports knowledge representation (domain ontologies and task structures) and generates natural language questions to be used as issues for the debate. This work also describes the software prototype that implements the method and the experimental results frOID applying AMANDA in actual training situations
N'dongo, Harouna. "Environnements personnels d'apprentissage : modélisation multi-agents d'un support de construction". Master's thesis, Université Laval, 2018. http://hdl.handle.net/20.500.11794/32467.
Texto completo da fonteWith Web 2.0, the university community have developed new learning practices. Students adopt practices for integrating their social activities into their learning process. The needs of learners are rapidly changed and institutions do not still offer resources totally adapted to this context. Students expect to have access to a vast amount of information through formal or informal tools and to be part of a global peer community. The need to combine institutional and personal resources implies a new concept: Personal Learning Environment (PLE). A PLE is defined as a space for learners to customize the aggregation of heterogeneous resources for their learning needs. Developing such environments requires the involvement of complex skills that higher education students, though described as digital natives, do not necessarily possess. The objective of this research work is to allow students to have support for the construction of their PLE as a dashboard. It would provide an up - to - date view of resources use and further exploit the resulting linkages. The proposed solution is based on the Agent - Group - Role organizational model, one of design method for multi-agent systems. A prototype was also implemented and allowed to demonstrate the feasibility of the proposed solution.
Abdelghani, Rania. "Guider les esprits de demain : agents conversationnels pour entraîner la curiosité et la métacognition chez les jeunes apprenants". Electronic Thesis or Diss., Bordeaux, 2024. http://www.theses.fr/2024BORD0152.
Texto completo da fonteEpistemic curiosity—the desire to actively seek information for its inherent pleasure—is a complex phenomenon extensively studied across various domains. Several researchers in psychology, neuroscience, and computer science have repeatedly highlighted its foundational role in cognitive development and in fostering lifelong learning. Further, epistemic curiosity is considered key for cultivating a flexible mindset capable of adapting to the world’s uncertainties. These insights have spurred significant interest in the educational field, recognizing curiosity as essential for helping individuals be active and in control of their learning. These properties are crucial for addressing some of today’s major educational challenges, namely offering students individualized support to suit their competencies and motivations, and helping them become able to learn autonomously and independently in their dynamic and uncertain environments. Despite this well-documented importance of curiosity in education, its practical implementation and promotion in the classroom remains limited. Notably, one of the primary expressions of curiosity— question-asking (QA)—is nearly absent in most of today’s educational settings. Several reports show that students often spend a lot of time answering teachers’ questions rather than asking their own. And when they do ask questions, they are typically low-level and memory-based, as opposed to curious questions that seek novel information. In this context, this thesis aims to develop educational technologies that can foster children’s curiosity-driven learning by practicing curious QA behaviors, and their related metacognitive (MC) skills. Ultimately, we implemented interventions to train three dimensions: 1) Linguistic QA Skills: We implement a conversational agent to train the ability to formulate curious questions using compound questioning words and correct interrogative constructions. It helps children generate curious questions during reading-comprehension tasks, by providing specific cues. The effectiveness of different cue structures (a sentence vs. series of keywords) and implementations (hand-generated vs. GPT-3-generated content) is studied. 2) Curiosity-related metacognitive Skills: We create animated videos to give declarative knowledge about curiosity and its related MC skills: the ability to self reflect, make educated guesses, formulate efficient questions, and evaluate newly-acquired information. We also propose sessions to practice these skills during reading-comprehension tasks using specific cues given by conversational agents we designed to train procedural MC. 3) Social Perceptions and beliefs: We create animated videos to address the negative constructs learners tend to have about curiosity. They explain the importance of curiosity and how to control it during learning. Over 150 French students aged 9 to 11 were recruited to test these trainings of the three dimensions. Combined, these latter enhanced students’ MC sensitivity and perception of curiosity. At their turn, these factors facilitated students’ divergent QA behaviors which, at their turn, led to stronger learning progress and positive, affordable learning experiences. But despite the positive results, our methods had limitations, particularly their short duration. We suggest testing longer-lasting interventions to examine their long-term effects on curiosity. Finally, this thesis highlights the need to continue exploring QA and MC research in the age of Generative Artificial Intelligence (GAI). Indeed, while GAI facilitates access to information, it still requires good QA abilities and MC monitoring to prevent misinformation and facilitate its detection. We thus propose a framework to link efficient GAI use in education to QA and MC skills, and GAI literacy. We also present a behavioral study we intend to conduct to test this framework
Ciguene, Richardson. "Génération automatique de sujets d'évaluation individuels en contexte universitaire". Electronic Thesis or Diss., Amiens, 2019. http://www.theses.fr/2019AMIE0046.
Texto completo da fonteThis PhD work focuses on the evaluation of learning and especially the automatic generation of evaluation topics in universities. We rely on a base of source questions to create topic questions through algorithms that are able to construct differentiated assessment tests. This research has made it possible to develop a metric that measures this differentiation and to propose algorithms aimed at maximizing total differentiation on test collections, while minimizing the number of necessary patterns. The average performance of the latter depends on the number of patterns available in the source database (compared to the number of items desired in the tests), and the size of the generated collections. We focused on the possible differentiation in very small collections of subjects, and proposes methodological tracks to optimize the distribution of these differentiated subjects to cohorts of students respecting the constraints of the teacher. The rest of this work will eventually take into account the level of difficulty of a test as a new constraint, relying in part on the statistical and semantic data collected after each test. The goal is to be able to maximize the differentiation by keeping the equity between the Tests of a Collection, for an optimized distribution during the Events
Bouchereau, Aymeric. "Les objets connectés au service de l'apprentissage". Thesis, Bourgogne Franche-Comté, 2020. http://www.theses.fr/2020UBFCC018.
Texto completo da fonteLearning - a vital principle of evolution - ensures the transformation of primary data captured by our senses into useful knowledge or abstract and general ideas that can be used in new situations and contexts. Cognitive neuroscience shows that the mechanisms of learning are stimulated by cognitive (e.g. wondering, evaluating errors), physical (e.g. manipulating, moving) and social (e.g. debating, collaborating) engagement. The learner builds knowledge through experience, by exploring his environment, formulating hypotheses and experimenting.Learning is crucial in a context where the exponential evolution of information and communication technologies is changing objects, practices and uses. The development of the Internet of Things (IoT) transforms common objects (e.g. light bulbs, watches, cars) into connected devices (CD) that can collect data and act on the user's environment. Learning becomes both biological and artificial and allows the creation of artificial intelligence systems (AIS) that analyse large volumes of data to automate tasks and assist individuals.Technologies can support learning when the technical possibilities they offer are used to support the process of knowledge construction. Thus, this thesis focuses on learning in the context of IoT and examines how the specificities of CD can be articulated with the mechanisms of learning.In order to identify the characteristics of learning in the context of IoT, we studied existing uses of CD. Based on the state of the art, we proposed a conceptual tool describing the IoT through four dimensions of analysis: Data, Interfaces, Agents and Pervasiveness. This tool enabled us to identify, list, classify and ultimately analyse the uses of CD for learning. In the context of these uses, learning is characterised by physical commitment, contextualisation of knowledge and bringing pedagogical activities closer to reality.Building on the results of this initial work, we have developed an approach to put the specificities of CD to learn sciences. The abstract and often counter-intuitive aspect of scientific knowledge hinders their learning, partly because our perception of reality is subjective and limited by our senses. However, data collected by CD and analysed by AIS provide information about the environment that can be used to extend human perception.Therefore, the objective of our approach, translated by the Data - Representations - Interactions (DRI) model, aims at exploiting OCs and SIAs to facilitate the observation of physical phenomena. According to the DRI model, the learner interacts with representations of a physical phenomenon generated by CD and AIS. In accordance with the mechanisms of learning (e.g. constructivism, role of experience), the learner is led to make observations and manipulations, formulate hypotheses and test them. In order to evaluate the effects and constraints of the DRI model, we have designed LumIoT devices dedicated to the learning of photometric quantities (e.g. luminous flux, luminous intensity, illuminance). Then, we conducted an experiment with 17 students of the Master 1 Multimedia Products and Services of the University of Franche-Comté (Montbéliard).The results of the experiment show that the LumIoT devices, based on the DRI model, have facilitated the observation and understanding of photometric quantities. By making abstract knowledge accessible, the DRI model paves the way for learning devices using CD and AIS to mediate knowledge
Popineau, Fabrice. "Approche Logique de la Personnalisation dans les Environnements Informatiques pour l'Apprentissage Humain". Electronic Thesis or Diss., université Paris-Saclay, 2023. http://www.theses.fr/2023UPASG001.
Texto completo da fonteWe present here a body of works dealing with learner adaptation in Technology Enhanced Learning (TLE) platforms. These works are based on a classical approach of artificial intelligence, essentially grounded on logic and an agent approach. We develop several arguments on the appropriateness of situation calculus to drive these platforms. We propose a method of transforming logic programs to adapt the agent's behavior to the user, or to give a personality to the agent. We also discuss the possibility of integrating this approach in a MOOC platform, to accompany learners and to provide them with personalized recommendations
Quanquin, Véronique. "Le choix du texte : problématique et application dans le cadre d'un environnement informatique d'aide à l'enseignement et l'apprentissage de la lecture en cours préparatoire". Clermont-Ferrand 2, 2000. http://www.theses.fr/2000CLF20011.
Texto completo da fonteGharroudi, Ouadie. "Ensemble multi-label learning in supervised and semi-supervised settings". Thesis, Lyon, 2017. http://www.theses.fr/2017LYSE1333/document.
Texto completo da fonteMulti-label learning is a specific supervised learning problem where each instance can be associated with multiple target labels simultaneously. Multi-label learning is ubiquitous in machine learning and arises naturally in many real-world applications such as document classification, automatic music tagging and image annotation. In this thesis, we formulate the multi-label learning as an ensemble learning problem in order to provide satisfactory solutions for both the multi-label classification and the feature selection tasks, while being consistent with respect to any type of objective loss function. We first discuss why the state-of-the art single multi-label algorithms using an effective committee of multi-label models suffer from certain practical drawbacks. We then propose a novel strategy to build and aggregate k-labelsets based committee in the context of ensemble multi-label classification. We then analyze the effect of the aggregation step within ensemble multi-label approaches in depth and investigate how this aggregation impacts the prediction performances with respect to the objective multi-label loss metric. We then address the specific problem of identifying relevant subsets of features - among potentially irrelevant and redundant features - in the multi-label context based on the ensemble paradigm. Three wrapper multi-label feature selection methods based on the Random Forest paradigm are proposed. These methods differ in the way they consider label dependence within the feature selection process. Finally, we extend the multi-label classification and feature selection problems to the semi-supervised setting and consider the situation where only few labelled instances are available. We propose a new semi-supervised multi-label feature selection approach based on the ensemble paradigm. The proposed model combines ideas from co-training and multi-label k-labelsets committee construction in tandem with an inner out-of-bag label feature importance evaluation. Satisfactorily tested on several benchmark data, the approaches developed in this thesis show promise for a variety of applications in supervised and semi-supervised multi-label learning
Enguehard, Chantal. "Acquisition naturelle automatique d'un réseau sémantique". Compiègne, 1992. http://www.theses.fr/1992COMPD527.
Texto completo da fonteMartin, Bruno. "Diagnostic comportemental et cognitif des erreurs dans la résolution de problèmes arithmétiques". Thesis, Paris 8, 2016. http://www.theses.fr/2016PA080084/document.
Texto completo da fonteResearch in psychology and in Intelligent Learning Environment (ILE) share the goal of a better understanding of the subject, more precisely in its modeling with strong educative perspectives. This Ph.D. Thesis, intrinsically interdisciplinary, aims to strengthen the link between experimental psychology and ILEs, especially in the domain of arithmetical word problem solving (WP) while addressing the more general issue of the integration of cognitive models in the ILEs. In the first part, a behavioral diagnostic module is presented, with a test-based methodology to assess its relevance in the context of experimental psychology studies. A better understanding of word problem solving behavior is a prerequisite of the development of any cognitive diagnostic module. This is the core of the second part, which presents the cognitive models put in place and compare their productions with human data. It has been shown that a large part of WP solving behavior can be explained via the light of keyword-based strategy and the alteration of the meaning of textual propositions. In the last part, in order to address the issue of individual cognitive diagnosis, a metric quantifying the fit of the diagnosis has been developed. A software has also been developed, allowing to build and test simple cognitive models over data coming from ILEs. This metric and this software have been used concretely within the context of an experimentation involving word problem solving
Ergün, Mustafa. "Etude didactique d'une activité de conception de protocole expérimental en chimie, supportée par un environnement informatique dédié : étude de cas d'Educ@ffix.net". Grenoble 1, 2006. http://www.theses.fr/2006GRE10057.
Texto completo da fonteThis thesis highlights the necessary knowledge and learning acquired during an experimental design task in chemistry with the assistance of an ILE (Intelligent Learning Environment). We focused on the capacity of last year of high school pupils to design an experiment, by using the ILE named Educ@ffix. Net, which answers the following problem: “determine the concentration of a substance with a spectrophotometric method”. We attempted to identify the nature of the difficulties met by the pupils, in order to distinguish those related to the discipline from those related to the context. We also studied the chosen strategies as well as the role attributed to the artificial tutor by the pupils during their task. The thesis also studies the acquisition related to the experimental design task with the dedicated ILE. For this propose, we located theoretical and procedural knowledge necessary for the pupils in their task and highlighted the relations performed between a procedural knowledge and an associated theoretical knowledge. The obtained results underline that (1) the studied ILE enables the pupils to achieve the required task, (2) misconceptions of the pupils and the previous didactical contract involve difficulties for them during the experimental design task, (3) theoretical knowledge necessary to the construction of procedural knowledge is not always controlled by the pupils, (4) theoretical course available for the pupils in this ILE and the artificial tutor feedbacks often lead to a discussion within the pair which helps in the success of the task and in the build of procedural knowledge
Oulhaci, M'hammed Ali. "Évaluation individuelle et collective dans les jeux sérieux collaboratifs : application à la gestion de crise". Thesis, Aix-Marseille, 2014. http://www.theses.fr/2014AIXM4314/document.
Texto completo da fonteSerious Games (SG) are increasingly used in the field of training, SG aims to teach one or more skills(s) related to a business or to more general qualities, to one or more learners (players), this with a playful dimension. Usually, SG aims a specific training objective related to a specific context training which can be linked to a trade (eg fire ), a body of knowledge (school or university), or the acquisition of social skills (conflict management, cooperation ...). To Measure the success or failure of this learning is ultimately an assessment process of the learner(s) and in recent years became a major issue in SG in addition to issues classic design and screenwriting.In the context of multi-player SG, with a collaborative tasks learning performed by different learners (or virtual actor) sharing a common goal, our research focuses on the assessment of learning processes, leading to both individual and collective assessment. We propose first a methodological framework for learners assessment. This methodological framework is based on the "Evaluation space" concept associated with a specific view corresponding to a particular assessment objective, leading to the calculation of indicators, based on specific knowledge models representation and assessment. In order to operationalize this methodological assessment framework, we propose a agents based architecture, on the one hand allowing human learners and virtual players (simulation of behavior) interaction, and other hand supporting individual and collective learners' assessment. This architecture is implemented and integrated to the SIMFOR SG, and illustrated in a crisis management scenario exercise
Renié, Delphine. "Modélisation informatique de l'acquisition des interrogatives directes en français langue seconde dans leur dimension pragmatique, proposition d'un environnement offrant un apprentissage collaboratif : eleonore". Clermont-Ferrand 2, 1995. http://www.theses.fr/1995CLF20080.
Texto completo da fonteThe elaboration of the computer environment eleonore has aimed at two main goals : to support learning on the one hand, and to propose a way modelling the acquisition of one aspect of the second language on the other hand. Our study starts with a description of the object of learning : direct interrogatives in french. These propositions are an example of lingistic constructions for which syntax, semantics and pragmatics cannot be isolated if one is to consider the behaviour of a native speaker of french in a communicative situation. That behaviour is simulated by a generator, in our computer environment. The acquisition of interrogative sentences in a first and in a second language is then considered, in their morpho-syntactic and their pragmatic aspects. The data on the acquisition of pragmatics in a second language which we have collected through a longitudinal experimentation with learners of french, are a complement to this study. It is only after we have integrated them, that we are able to suggest a model of the acquisition of interrogative sentences. Moreover, the data enable us to determine the pedagogical founding of eleonore, and to integrate a learner model in the architecture of our environment. The model acts all through a working session with eleonore, especially in the last activity which offers a collaboration between the learner and a simulated companion. Indeed, eleonore includes several types of activities based either on induction, deduction or on both. The activities can concern pragmatic aspects of interrogatives, syntactic ones or both. For an activity involving all aspects, we preconise a collaborative learning mode in which a simulated learner acts as a native speaker with whom the learner is in a situation of exolingual interaction. The environment is then evaluated through a formative evaluation concerning the environment itself rather than the learning it supports. The refinement of our acquisition model through the recourse to a simulated learner constitutes the next orientation in the development of eleonore, as detailed in the last chapter
Di, Marco Lionel. "Récit d'ingénierie pédagogique en santé à l'usage de l'enseignant connecté Does the acceptance of hybrid learning affect learning approaches in France? Blended Learning for French Health Students: Does Acceptance of a Learning Management System Influence Students’ Self-Efficacy?" Thesis, Université Grenoble Alpes, 2020. http://www.theses.fr/2020GRALS024.
Texto completo da fonteBackground. The general objective of this thesis was to evaluate a hybrid pedagogical method using an integrated learning environment (ILE) in the training of health professionals. Three research questions followed one after the other. Does the acceptability of blended learning affect students' learning strategies and learning approaches? Does the acceptability of an ILE affect students' self-efficacy? What characteristics of a dematerialised course make students' attention variable?Materials & Methods. We carried out 2 quantitative observational studies, as well as a single-blind observational experiment coupled with a qualitative analysis, with different classes of midwifery students of Grenoble-Alpes University Faculty of Medicine.Results. Students have suited learning approaches and strategies despite the use of a hybrid teaching method which they reject; there is no correlation between poor acceptability of the ILE and different spheres of students' self-efficacy; and the variability of attention declared by students varies according to certain factors common to those detected by artificial intelligence (type of language, slide duration…).Discussion. The internal and external validities of this work highlight the close links between interest, motivation, engagement by identification, and attention. It is thus possible to put forward principles of pedagogical engineering in health within the framework of dematerialized courses focusing on the content, form and type of knowledge capsule. Finally, the health teacher must above all be “connected to” the students, so that technical developments can be adapted to their needs
Cléder, Catherine. "Planification didactique et construction de l'objectif d'une session de travail individualisée : modélisation des connaissances et du raisonnement mis en jeu". Clermont-Ferrand 2, 2002. http://www.theses.fr/2002CLF20019.
Texto completo da fonteLodter, Christiane. "Intelligence artificielle en orthodontie (ODONTEL*)". Toulouse 3, 1990. http://www.theses.fr/1990TOU30206.
Texto completo da fonteBessière, Christian. "Systèmes à contraintes évolutifs en intelligence artificielle". Montpellier 2, 1992. http://www.theses.fr/1992MON20166.
Texto completo da fonteMachrouh, Joseph. "Perception attentive et vision en intelligence artificielle". Paris 11, 2002. http://www.theses.fr/2002PA112301.
Texto completo da fonteThe work presented in this thesis deals with the development of software agents endowed with perceptive capacities. To provide such a system with exploratory capacities supposes the determination of interest points in the scene. In order to be able to move in the image, one will distinguish a low-resolution wide field processing and a high resolution focal processing. One thus separates the exploration phase associated to the search of interest points from the exploitation phase associated to recognition recognition. The selected points consist of energy maxima computed using wavelet filters covering a range of orientations and frequencies. The low frequencies are used to determine the peripheral saliency. Principal Component Analysis (PCA) projection system was computed from a representative sample of natural scenes. This system was used to categorize the interest points of an unspecified scene. The system thus can use several points of view to guide its attentionnal mechanisms. The energy of these interest points according to various orientations and space frequencies is then used to index them. We showed that the low frequency components of this indexed representation are sufficient to bias the saliency of the scene in favor of targets similar to the representations memorized. They are also sufficiently robust to preserve this property in a video sequence subject to strong contrast variations. We showed as well that interest points based on a multi-scale frequency analysis can be used to control exploratory saccades by using a bottom-up mechanism; the low frequency part of such a representation can be used to control the saccades required to attain the target in a top-down way
Cornu, Philippe. "Logiques et sciences cognitives en intelligence artificielle". Grenoble 2 : ANRT, 1987. http://catalogue.bnf.fr/ark:/12148/cb37604052k.
Texto completo da fonteHelft, Nicolas. "L'Induction en intelligence artificielle théorie et algorithmes /". Grenoble 2 : ANRT, 1988. http://catalogue.bnf.fr/ark:/12148/cb376141929.
Texto completo da fonteAlliche, Abderrahmane Redha. "Contrôle du réseau cloud basé intelligence artificielle". Electronic Thesis or Diss., Université Côte d'Azur, 2024. http://www.theses.fr/2024COAZ4022.
Texto completo da fonteThe exponential growth of Internet traffic in recent decades has prompted the emergence of Content Delivery Networks (CDNs) as a solution for managing high traffic volumes through data caching in cloud servers located near end-users. However, challenges persist, particularly for non-cacheable services, necessitating the use of cloud overlay networks. Due to a lack of knowledge about the underlay network, cloud overlay networks introduce complexities such as Triangle inequality violations (TIV) and dynamic traffic routing challenges.Leveraging the Software Defined Networks (SDN) paradigm, Deep Reinforcement Learning (DRL) techniques offer the possibility to exploit collected data to better adapt to network changes. Furthermore, the increase of cloud edge servers presents scalability challenges, motivating the exploration of Multi-Agent DRL (MA-DRL) solutions. Despite its suitability for the distributed packet routing problem in cloud overlay networks, MA-DRL faces non-addressed challenges such as the need for realistic network simulators, handling communication overhead, and addressing the multi-objective nature of the routing problem.This Ph.D. thesis delves into the realm of distributed Multi-Agent Deep Reinforcement Learning (MA-DRL) methods, specifically targeting the Distributed Packet Routing problem in cloud overlay networks. Throughout the thesis, we address these challenges by developing realistic network simulators, studying communication overhead in the non-overlay general setting, and proposing a distributed MA-DRL framework tailored to cloud overlay networks, focusing on communication overhead, convergence, and model stability
Aouag, Sofiane. "Individualisation de l'apprentissage dans un Système Tuteur Intelligent : cas de l'apprentissage de la lecture dans un système AMICAL". Phd thesis, Université Blaise Pascal - Clermont-Ferrand II, 2010. http://tel.archives-ouvertes.fr/tel-00658846.
Texto completo da fonteGuérin, Marianne Penit. "Intelligence artificielle et électrophysiologie sensorielle oculaire : Pantops PC". Clermont-Ferrand 1, 1987. http://www.theses.fr/1987CLF11037.
Texto completo da fonteEl, Alam Iyad. "Management des compétences : nouvelles technologies et intelligence artificielle". Aix-Marseille 3, 2007. http://www.theses.fr/2007AIX32077.
Texto completo da fonteThe study focuses on the systemic and mathematical modelisation of the skills adjustment problems, both in the classical approaches of operational research models and in new viewpoints stemming from neural networks and artificial intelligence. This modelisation is conducted within the framework of a decision support system of evaluation and skills match through the contribution of the modelisation of micro- and macro-competencies. The context of artificial intelligence will be that of multilayer neural networks, and of the so-called Fuzzy ART, with the aim of proposing a system which we have called CRMM (Competencies Research Matching Model). The system will be implemented in large-sized organizations in need of large numbers of personnel subject to frequent post changes. The assumption on which this study is based tends to demonstrate the possibility to improve the decision-making of human resources or operational managers through a better exploitation of competency-related data. This factor should obviously be placed at the core of new strategic problematics linked to competencies and training
Pouget, Jonathan. "La réparation du dommage impliquant une intelligence artificielle". Thesis, Aix-Marseille, 2019. http://theses.univ-amu.fr.lama.univ-amu.fr/191212_POUGET_871qcngtj900zlfid640gfeuf393zytl_TH.pdf.
Texto completo da fonteDo the current tort law and insurance law allow an adequate compensation for the damage caused, directly or indirectly, by anartificial intelligence? Firstly, this thesis demonstrates that tort law and insurance law need to be interprated. Secondly, it demonstrates that a part of tort law and insurance contracts have to evolve. The track of a legal personhood with supervised functions and attributable to artificial intelligences will also be studied for these purposes
Le, Gauffre Pascal. "Méthodologie de conception et intelligence artificielle en bâtiment". Lyon, INSA, 1988. http://www.theses.fr/1988ISAL0045.
Texto completo da fonteResearch on Computer Aided Design requires a double conceptual base. On the one hand, we have to represent what is the design process, on the other hand we have to define the possible and desirable contributions of the computer in this design process. The works presented deal with Computer Aided Building Design and the contribution of Artificial Intelligence techniques. The first part of this report is dedicated to the study of conceptual bases for Computer Aided Building Design and refers to Information Processing System theory as well as to systems modelisation. The design process is regarded as a modelling activity and as a problem setting (and solving). Process. The contribution of Artificial Intelligence techniques is therefore considered for the representation of the different expert considerations on the building project and for the construction and the resolution of problems. In the second part of the report we present a prototype of an expert system, called DEDALE, dedicated to the four following tasks : 1- Modelisation of the building project, 2- Selection of a strategy for modifications, 3- Construction of a restricted multi criteria problem (according to the strategy), 4- Management of the resolution process. Finally, a third part is devoted to a presentation of an experiment of the DEDALE system, and to a presentation of future prospects of this tool and of the approach used to elaborate it
Helft, Nicolas. "L' Induction en intelligence artificielle : théorie et algorithmes". Aix-Marseille 2, 1988. http://www.theses.fr/1988AIX22044.
Texto completo da fonteDurand, Stéphane. "Représentation des points de vues multiples dans une situation d'urgence : une modélisation par organisations d'agents". Le Havre, 1999. http://www.theses.fr/1999LEHA0005.
Texto completo da fonteImbert, Jean-Louis. "Simplification des systèmes de contraintes numériques linéaires". Aix-Marseille 2, 1989. http://www.theses.fr/1989AIX22022.
Texto completo da fonteVenturini, Gilles. "Apprentissage adaptatif et apprentissage supervise par algorithme genetique". Paris 11, 1994. http://www.theses.fr/1994PA112016.
Texto completo da fontePistilli, Giada. "Pour une éthique de l'intelligence artificielle conversationnelle". Electronic Thesis or Diss., Sorbonne université, 2024. http://www.theses.fr/2024SORUL038.
Texto completo da fonteThis research aims to probe the ethical intricacies of conversational Artificial Intelligence (AI), specifically focusing on Large Language Models and conversational agents. This manuscript constructs a framework that melds empirical analysis with philosophical discourse. We aim to urgently advocate for a well-founded ethical structure for conversational AI, highlighting the necessity to involve all stakeholders, from developers to end-users. Firstly, we champion the integration of engineering and other scientific disciplines with philosophy, facilitating a more nuanced understanding of the ethical dimensions underpinning AI. This collaborative approach allows for a richer, more informed ethical discourse. Secondly, we advocate for the dynamic use of applied ethical frameworks as foundational guides for setting the initial objectives of an AI system. These frameworks serve as evolving tools that adapt to the ethical complexities encountered during development and deployment. Lastly, grounded in hands-on, interdisciplinary research, we make an argument for the prioritization of narrow, task-specific AI over Artificial General Intelligence, a stance that is based on the enhanced feasibility of ethical oversight and technical controllability.With this research, we aim to contribute to the literature on AI ethics, enriching the academic discourse in both philosophy and computer science