Dissertations / Theses on the topic 'Exploration interactive de données'
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 dissertations / theses for your research on the topic 'Exploration interactive de données.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse dissertations / theses on a wide variety of disciplines and organise your bibliography correctly.
Alam, Mehwish. "Découverte interactive de connaissances dans le web des données." Thesis, Université de Lorraine, 2015. http://www.theses.fr/2015LORR0158/document.
Full textRecently, the “Web of Documents” has become the “Web of Data”, i.e., the documents are annotated in the form of RDF making this human processable data directly processable by machines. This data can further be explored by the user using SPARQL queries. As web clustering engines provide classification of the results obtained by querying web of documents, a framework for providing classification over SPARQL query answers is also needed to make sense of what is contained in the data. Exploratory Data Mining focuses on providing an insight into the data. It also allows filtering of non-interesting parts of data by directly involving the domain expert in the process. This thesis contributes in aiding the user in exploring Linked Data with the help of exploratory data mining. We study three research directions, i.e., 1) Creating views over RDF graphs and allow user interaction over these views, 2) assessing the quality and completing RDF data and finally 3) simultaneous navigation/exploration over heterogeneous and multiple resources present on Linked Data. Firstly, we introduce a solution modifier i.e., View By to create views over RDF graphs by classifying SPARQL query answers with the help of Formal Concept Analysis. In order to navigate the obtained concept lattice and extract knowledge units, we develop a new tool called RV-Explorer (Rdf View eXplorer) which implements several navigational modes. However, this navigation/exploration reveal several incompletions in the data sets. In order to complete the data, we use association rule mining for completing RDF data. Furthermore, for providing navigation and exploration directly over RDF graphs along with background knowledge, RDF triples are clustered w.r.t. background knowledge and these clusters can then be navigated and interactively explored. Finally, it can be concluded that instead of providing direct exploration we use FCA as an aid for clustering RDF data and allow user to explore these clusters of data and enable the user to reduce his exploration space by interaction
Da, Costa David. "Visualisation et fouille interactive de données à base de points d'intérêts." Tours, 2007. http://www.theses.fr/2007TOUR4021.
Full textIn this thesis, we present the problem of the visual data mining. We generally notice that it is specific to the types of data and that it is necessary to spend a long time to analyze the results in order to obtain an answer on the aspect of data. In this thesis, we have developed an interactive visualization environment for data exploration using points of interest. This tool visualizes all types of data and is generic because it uses only one similarity measure. These methods must be able to deal with large data sets. We also sought to improve the performances of our visualization algorithms, thus we managed to represent one million data. We also extended our tool to the data clustering. Most existing data clustering methods work in an automatic way, the user is not implied iin the process. We try to involve more significantly the user role in the data clustering process in order to improve his comprehensibility of the data results
Hurter, Christophe. "Caractérisation de visualisations et exploration interactive de grandes quantités de données multidimensionnelles." Phd thesis, Université Paul Sabatier - Toulouse III, 2010. http://tel.archives-ouvertes.fr/tel-00610623.
Full textBen, Said Guefrech Zohra. "A virtual reality-based approach for interactive and visual mining of association rules." Nantes, 2012. http://archive.bu.univ-nantes.fr/pollux/show.action?id=359deab9-229a-4369-908d-bfbbe98adaea.
Full textThis thesis is at the intersection of two active research areas : Association Rules Mining and Virtual Reality. The main limitations of the association rule extraction algorithms are (i) the large amount of the generated rules and (ii) their low quality. Several solutions have been proposed to address this problem such as, the post-processing of association rules that allows rule validation and extraction of useful knowledge. Whereas rules are automatically extracted by combinatorial algorithms, rule post-processing is done by the user. Visualisation can help the user facing the large amount of rules by representing them in visual form. In order to find relevant knowledge in visual representations, the user needs to interact with these representations. To this aim, it is essential to provide the user with efficient interaction techniques. This work addresses two main issues : an association rule representation that allows the user quickly detection of the most interesting rules and interactive exploration of rules. The first issue requires an intuitive representation metaphor of association rules. The second requires an interactive exploration process allowing the user to explore the rule search space focusing on interesting rules. The main contributions of this work can be summarised as follows : – We propose a new classification for Visual Data Mining techniques, based on both 3D representations and interaction techniques. Such a classification helps the user choosing a visual representation and an interaction technique for his/her application. – We propose a new visualisation metaphor for association rules that takes into account the attributes of the rule, the contribution of each one, and their correlations. – We propose a methodology for interactive exploration of association rules to facilitate the user task facing large sets of rules taking into account his/her cognitive capabilities. In this methodology, local algorithms are used to recommend better rules based on a reference rule which is proposed by the user. Then, the user can both drives extraction and post-processing of rules using appropriate interaction operators. – We developed a tool that implements all the methodology functionality. The tool is based on an intuitive display in a virtual environment and supports multiple interaction methods
Djedaini, Mahfoud. "Automatic assessment of OLAP exploration quality." Thesis, Tours, 2017. http://www.theses.fr/2017TOUR4038/document.
Full textIn a Big Data context, traditional data analysis is becoming more and more tedious. Many approaches have been designed and developed to support analysts in their exploration tasks. However, there is no automatic, unified method for evaluating the quality of support for these different approaches. Current benchmarks focus mainly on the evaluation of systems in terms of temporal, energy or financial performance. In this thesis, we propose a model, based on supervised automatic leaming methods, to evaluate the quality of an OLAP exploration. We use this model to build an evaluation benchmark of exploration support sys.terns, the general principle of which is to allow these systems to generate explorations and then to evaluate them through the explorations they produce
Wang, Xiyao. "Augmented reality environments for the interactive exploration of 3D data." Electronic Thesis or Diss., université Paris-Saclay, 2020. http://www.theses.fr/2020UPASG052.
Full textExploratory visualization of 3D data is fundamental in many scientific domains. Traditionally, experts use a PC workstation and rely on mouse and keyboard to interactively adjust the view to observe the data. This setup provides immersion through interaction---users can precisely control the view and the parameters, but it does not provide any depth clues which can limit the comprehension of large and complex 3D data. Virtual or augmented reality (V/AR) setups, in contrast, provide visual immersion with stereoscopic views. Although their benefits have been proven, several limitations restrict their application to existing workflows, including high setup/maintenance needs, difficulties of precise control, and, more importantly, the separation from traditional analysis tools. To benefit from both sides, we thus investigated a hybrid setting combining an AR environment with a traditional PC to provide both interactive and visual immersions for 3D data exploration. We closely collaborated with particle physicists to understand their general working process and visualization requirements to motivate our design. First, building on our observations and discussions with physicists, we built up a prototype that supports fundamental tasks for exploring their datasets. This prototype treated the AR space as an extension to the PC screen and allowed users to freely interact with each using the mouse. Thus, experts could benefit from the visual immersion while using analysis tools on the PC. An observational study with 7 physicists in CERN validated the feasibility of such a hybrid setting, and confirmed the benefits. We also found that the large canvas of the AR and walking around to observe the data in AR had a great potential for data exploration. However, the design of mouse interaction in AR and the use of PC widgets in AR needed improvements. Second, based on the results of the first study, we decided against intensively using flat widgets in AR. But we wondered if using the mouse for navigating in AR is problematic compared to high degrees of freedom (DOFs) input, and then attempted to investigate if the match or mismatch of dimensionality between input and output devices play an important role in users’ performance. Results of user studies (that compared the performance of using mouse, space mouse, and tangible tablet paired with the screen or the AR space) did not show that the (mis-)match was important. We thus concluded that the dimensionality was not a critical point to consider, which suggested that users are free to choose any input that is suitable for a specific task. Moreover, our results suggested that the mouse was still an efficient tool compared to high DOFs input. We can therefore validate our design of keeping the mouse as the primary input for the hybrid setting, while other modalities should only serve as an addition for specific use cases. Next, to support the interaction and to keep the background information while users are walking around to observe the data in AR, we proposed to add a mobile device. We introduced a novel approach that augments tactile interaction with pressure sensing for 3D object manipulation/view navigation. Results showed that this method could efficiently improve the accuracy, with limited influence on completion time. We thus believe that it is useful for visualization purposes where a high accuracy is usually demanded. Finally, we summed up in this thesis all the findings we have and came up with an envisioned setup for a realistic data exploration scenario that makes use of a PC workstation, an AR headset, and a mobile device. The work presented in this thesis shows the potential of combining a PC workstation with AR environments to improve the process of 3D data exploration and confirms its feasibility, all of which will hopefully inspire future designs that seamlessly bring immersive visualization to existing scientific workflows
Destandau, Marie. "Path-Based Interactive Visual Exploration of Knowledge Graphs." Electronic Thesis or Diss., université Paris-Saclay, 2020. http://www.theses.fr/2020UPASG063.
Full textKnowledge Graphs facilitate the pooling and sharing of information from different domains. They rely on small units of information named triples that can be combined to form higher-level statements. Producing interactive visual interfaces to explore collections in Knowledge Graphs is a complex problem, mostly unresolved. In this thesis, I introduce the concept of path outlines to encode aggregate information relative to a chain of triples. I demonstrate 3 applications of the concept withthe design and implementation of 3 open source tools. S-Paths lets users browse meaningful overviews of collections; Path Outlines supports data producers in browsing the statements thatcan be produced from their data; and The Missing Path supports data producers in analysingincompleteness in their data. I show that the concept not only supports interactive visual interfaces for Knowledge Graphs but also helps better their quality
Vidal, Jules. "Progressivité en analyse topologique de données." Electronic Thesis or Diss., Sorbonne université, 2021. http://www.theses.fr/2021SORUS398.
Full textTopological Data Analysis (TDA) forms a collection of tools that enable the generic and efficient extraction of features in data. However, although most TDA algorithms have practicable asymptotic complexities, these methods are rarely interactive on real-life datasets, which limits their usability for interactive data analysis and visualization. In this thesis, we aimed at developing progressive methods for the TDA of scientific scalar data, that can be interrupted to swiftly provide a meaningful approximate output and that are able to refine it otherwise. First, we introduce two progressive algorithms for the computation of the critical points and the extremum-saddle persistence diagram of a scalar field. Next, we revisit this progressive framework to introduce an approximation algorithm for the persistence diagram of a scalar field, with strong guarantees on the related approximation error. Finally, in a effort to perform visual analysis of ensemble data, we present a novel progressive algorithm for the computation of the discrete Wasserstein barycenter of a set of persistence diagrams, a notoriously computationally intensive task. Our progressive approach enables the approximation of the barycenter within interactive times. We extend this method to a progressive, time-constraint, topological ensemble clustering algorithm
Lavallard, Anne. "Exploration interactive d'archives de forums : Le cas des jeux de rôle en ligne." Phd thesis, Université de Caen, 2008. http://tel.archives-ouvertes.fr/tel-00292617.
Full textCouturier, Olivier. "Contribution à la fouille de données : règles d'association et interactivité au sein d'un processus d'extraction de connaissances dans les données." Artois, 2005. http://www.theses.fr/2005ARTO0410.
Full textPastor, Julien. "Conception d'une légende interactive et forable pour le SOLAP." Thesis, Université Laval, 2004. http://www.theses.ulaval.ca/2004/21994/21994.pdf.
Full textPatin, Gaël. "Extraction interactive et non supervisée de lexique en chinois contemporain appliquée à la constitution de ressources linguistiques dans un domaine spécialisé." Paris, INALCO, 2013. http://www.theses.fr/2013INAL0003.
Full textThis thesis deals with lexical unit extraction in contemporary Chinese from a corpus of specialized texts. It addresses the task of Chinese lexicon extraction using techniques based on linguistic characteristics of the Chinese language. The thesis also discusses how to evaluate the extraction of a lexicon in an industrial environment. The first part of the thesis describes the context of the study. We focus on describing the linguistic concepts of vocabulary and lexical units, and we also give a description of the construction of lexical units in contemporary Chinese. We then make a inventory of the different techniques used by the scientific community to address the task of extracting a contemporary Chinese lexicon. We conclude this section by describing lexicon extraction practices in industry, and we propose a formalization of the criteria used by terminologists to select the relevant lexical units. The second part of this thesis deals with the description of a method for extracting Chinese contemporary lexicon and its evaluation. We introduce a new numerical unsupervised method based on structural features of the lexical unit in Chinese and syntactic features of Chinese. The method includes an optional module to interact with a user (i. E. Semi-automatic). In the section related to the evaluation, we first evaluate the potential of the method by comparing extraction results to a reference standard and a reference method. We then implement a more pragmatic evaluation of the method by measuring the gains using this method as opposed to manual lexicon extraction by terminologists. The results obtained by our method are better than those produced by the reference method on the reference standard. These results are encouraging, but need to be confirmed by a more comprehensive study. The pragmatic evaluation shows that the method does not significantly improve the productivity of terminologists but can extract different lexical units than those obtained manually
Weber, Jonathan. "Segmentation morphologique interactive pour la fouille de séquences vidéo." Phd thesis, Université de Strasbourg, 2011. http://tel.archives-ouvertes.fr/tel-00643585.
Full textFangseu, Badjio Edwige P. "Evaluation qualitative et guidage des utilisateurs en fouille visuelle de données." Lyon 2, 2005. http://theses.univ-lyon2.fr/documents/lyon2/2005/fangseubadjio_ep.
Full textThe research context of these works is the visual data mining domain and more precisely supervised data classification. Other related fields are: knowledge extraction in the data, machine learning, quality of interface, software ergonomic, software engineering and human machine interaction. The result provided by a visual data mining tool is a data model. Generally, in order to access the quality of visual data mining tools, there is an estimation of the rate of bad classification. We believe that, this estimation is necessary but not sufficient for the evaluation of visual data mining tools. In fact, this type of tools use interfaces, graphical representations, data sets and require the participation of the end-users. On the basis of a state of the art on visualization, visual data mining and software quality, we propose two analysis and evaluation methods: an inspection method for experts and a diagnosis method which can be used by end-users for analysis and quality evaluation that takes account of the specificities of the treated domain. We developed guidelines and quality criteria (measures and metrics) for the analysis and the diagnosis of the visual data mining tools. From the users' point of view, in order to use information relating to their profiles and their preferences throughout the mining process, we also proposed a user model of visual data mining tools. Case studies performed with the proposed diagnosis method enable us to raise other problems than those resulting from the estimation of the rate of bad classification. This work presents also solutions brought to two problems listed during the analysis and the diagnosis of some existing visual data mining tools: the choice of the best algorithm to perform for a supervised classification task and the pre-treatment of very large data sets. We considered the problem of the choice of the best classification algorithm as a multi criteria decision problem. Artificial intelligence allows bringing solutions to the multi criteria analysis. We use the results coming from this domain through the multi-agents paradigm and the case based reasoning to propose a list of algorithms of decreasing effectiveness for the resolution of a given problem and to evolve knowledge of the case base. For the treatment of very large data sets, the limits of visual approaches concerning the number of records and the number of attributes are known. To be able to treat these data sets, a solution is to perform a pre-treatment of the data set before applying the interactive algorithm. The reduction of the number of records is performed by the application of a clustering algorithm, the reduction of the number of attributes is done by the combination of the results of feature selection algorithms by applying the consensus theory (with a visual weight assignment tool). We evaluate the performances of our new approaches on data sets of the UCI and the Kent Ridge Bio Medical Dataset Repository
Rayar, Frédéric. "Exploration interactive, incrémentale et multi-niveau de larges collections d'images." Thesis, Tours, 2016. http://www.theses.fr/2016TOUR4012/document.
Full textThe research work that is presented and discussed in this thesis focuses on large and evergrowing image collections. More specifically, we aim at providing one the possibility to explore such image collections, either to extract some kind of information and knowledge, or to wander in the collections. This thesis addresses this issue from the perspective of Interactive Data Exploration and Analytics. We take advantage of the similarity-based image collection browsing paradigm and aim at meeting simultaneously the three following constraints: (i) handling large image collections, up to millions of images, (ii) handling dynamic image collections, to deal with ever-growing image collections, and (iii) providing interactive means to explore image collections. To do so, we jointly study the indexing and the interactive visualisation of large and ever-growing image collections
Triperina, Evangelia. "Visual interactive knowledge management for multicriteria decision making and ranking in linked open data environments." Thesis, Limoges, 2020. http://www.theses.fr/2020LIMO0010.
Full textThe dissertation herein involves research in the field of the visual representations aided by semantic technologies and ontologies in order to support decisions and policy making procedures, in the framework of research and academic information systems. The visualizations will be also supported by data mining and knowledge extraction processes in the linked data environment. To elaborate, visual analytics’ techniques will be employed for the organization of the visualizations in order to present the information in such a way that will utilize the human perceptual abilities and that will eventually assist the decision support and policy making procedures. Furthermore, the visual representation and consequently the decision and policy making processes will be ameliorated by the means of the semantic technologies based on conceptual models in the form of ontologies. Thus, the main objective of the proposed doctoral thesis consists the combination of the key semantic technologies with interactive visualisations techniques based mainly on graph’s perception in order to make decision support systems more effective. The application field will be the research and academic information systems
El, Moussawi Adnan. "Clustering exploratoire pour la segmentation de données clients." Thesis, Tours, 2018. http://www.theses.fr/2018TOUR4010/document.
Full textThe research work presented in this thesis focuses on the exploration of the multiplicity of clustering solutions. The goal is to provide to marketing experts an interactive tool for exploring customer data that considers expert preferences on the space of attributes. We first give the definition of an exploratory clustering system. Then, we propose a new semi-supervised clustering method that considers user’s quantitative preferences on the analysis attributes and manages the sensitivity to these preferences. Our method takes advantage of metric learning to find a compromise solution that is both well adapted to the data structure and consistent with the expert’s preferences. Finally, we propose a prototype of exploratory clustering for customer relationship data segmentation that integrates the proposed method. The prototype also integrates visual and interaction components essential for the implementation of the exploratory clustering process
Rajaonarivo, Hiary Landy. "Approche co-évolutive humain-système pour l'exploration de bases de données." Thesis, Brest, 2018. http://www.theses.fr/2018BRES0114/document.
Full textThis thesis focus on a proposition that helps humans during the exploration of database. The particularity of this proposition relies on a co-evolution principle between the user and an intelligent interface. It provides a support to the understanding of the domain represented by the data. A metaphor of living virtual museum is adopted. This museum evolves incrementally according to the user's interactions. It incarnates both the data and the semantic information which are expressed by a knowledge model specific to the domain of the data. Through the topological organization and the incremental evolution, the museum personalizes online the user's exploration. The approach is insured by three main mechanisms: the evaluation of the user profile modelled by a dynamical weighting of the semantic information, the use of this dynamic profile to establish a recommendation as well as the incarnation of the data in the living museum. The approach has been applied to the heritage domain as part of the ANTIMOINE project, funded by the National Research Agency (ANR). The genericity of the latter has been demonstrated through its application to a database of publications but also using various types of interfaces (website, virtual reality).Experiments have validated the hypothesis that our system adapts itself to the user behavior and that it is able, in turn, to influence him.They also showed the comparison between a 2D interface and a 3D interface in terms of quality of perception, guidance, preference and efficiency
Guettala, Abdelheq Et-Tahir. "VizAssist : un assistant utilisateur pour le choix et le paramétrage des méthodes de fouille visuelle de données." Thesis, Tours, 2013. http://www.theses.fr/2013TOUR4017/document.
Full textIn this thesis, we deal with the problem of automating the process of choosing an appropriate visualization and its parameters in the context of visual data mining. To solve this problem, we developed a user assistant "VizAssist" which mainly assist users (experts and novices) during the process of exploration and analysis of their dataset. We illustrate the approach used by VizAssit to help users in the visualization selection and parameterization process. VizAssist proposes a process based on two steps. In the first step, VizAssist collects the user’s objectives and the description of his dataset, and then proposes a subset of candidate visualizations to represent them. In this step, VizAssist suggests a different mapping between the database for representation and the set of visualizations it manages. The second step allows user to adjust the different mappings suggested by the system. In this step, VizAssist uses an interactive genetic algorithm to allow users to visually evaluate and adjust such mappings. We present finally the results that we have obtained during the user evaluation that we performed and the contributions of our tool to accomplish some tasks of data mining
Ben, Said Zohra. "A virtual reality-based approach for interactive and visual mining of association rules." Phd thesis, Université de Nantes, 2012. http://tel.archives-ouvertes.fr/tel-00829419.
Full textFérey, Nicolas. "Exploration immersive de données génomiques textuelles et factuelles : vers une approche par visual mining." Paris 11, 2006. http://www.theses.fr/2006PA112235.
Full textThis thesis concerns the immersive exploration of textual and factual genomic data. The goal of this work is to design and study new approach for exploring genomic data within an immersive framework (i. E. Of virtual reality). The knowledge about genome is constituted by factual data, coming from structured biological or genomic databanks, and by textual data, namely the unstructured data within the millions publications relating to the research about genome. These data are heterogeneous, huge in quantity, and complex. The stake of this work is to propose visualization and interaction paradigms, which are able to deals with these characteristics. These paradigms must also be adapted to the immersive framework, and must respect the needs of the biologists. We used common points of genomic databanks, to design an original visualization paradigm, where the user is able to choice a translation of the semantic of the genomic data to visual, geometric or topologic properties. We implemented a software prototype in order to test and validate the visualization paradigm within an immersive framework. In this context, we proposed and tested new interaction paradigms, in order to navigate, search and edit the genomic data during the immersive exploration. We used finally this software to lead several experiments of genomic data analysis with biologists, in order to measure the relevance of this visual mining approach on different kinds of genomic data
Omidvar, Tehrani Behrooz. "Optimization-based User Group Management : Discovery, Analysis, Recommendation." Thesis, Université Grenoble Alpes (ComUE), 2015. http://www.theses.fr/2015GREAM038/document.
Full textUser data is becoming increasingly available in multiple domains ranging from phone usage traces to data on the social Web. User data is a special type of data that is described by user demographics (e.g., age, gender, occupation, etc.) and user activities (e.g., rating, voting, watching a movie, etc.) The analysis of user data is appealing to scientists who work on population studies, online marketing, recommendations, and large-scale data analytics. However, analysis tools for user data is still lacking.In this thesis, we believe there exists a unique opportunity to analyze user data in the form of user groups. This is in contrast with individual user analysis and also statistical analysis on the whole population. A group is defined as set of users whose members have either common demographics or common activities. Group-level analysis reduces the amount of sparsity and noise in data and leads to new insights. In this thesis, we propose a user group management framework consisting of following components: user group discovery, analysis and recommendation.The very first step in our framework is group discovery, i.e., given raw user data, obtain user groups by optimizing one or more quality dimensions. The second component (i.e., analysis) is necessary to tackle the problem of information overload: the output of a user group discovery step often contains millions of user groups. It is a tedious task for an analyst to skim over all produced groups. Thus we need analysis tools to provide valuable insights in this huge space of user groups. The final question in the framework is how to use the found groups. In this thesis, we investigate one of these applications, i.e., user group recommendation, by considering affinities between group members.All our contributions of the proposed framework are evaluated using an extensive set of experiments both for quality and performance
Richard, Jérémy. "De la capture de trajectoires de visiteurs vers l’analyse interactive de comportement après enrichissement sémantique." Electronic Thesis or Diss., La Rochelle, 2023. http://www.theses.fr/2023LAROS012.
Full textThis thesis focuses on the behavioral study of tourist activity using a generic and interactive analysis approach. The developed analytical process concerns the tourist trajectory in the city and museums as the study field. Experiments were conducted to collect movement data in the tourist city using GPS signals, thus enabling the acquisition of a movement trajectory. However, the study primarily focuses on reconstructing a visitor’s trajectory in museums using indoor positioning equipment, i.e., in a constrained environment. Then, a generic multi-aspect semantic enrichment model is developed to supplement an individual’s trajectory using multiple context data such as the names of neighborhoods the individual passed through in the city, museum rooms, weather outside, and indoor mobile application data. The enriched trajectories, called semantic trajectories, are then analyzed using formal concept analysis and the GALACTIC platform, which enables the analysis of complex and heterogeneous data structures as a hierarchy of subgroups of individuals sharing common behaviors. Finally, attention is paid to the "ReducedContextCompletion" algorithm that allows for interactive navigation in a lattice of concepts, allowing the data analyst to focus on the aspects of the data they wish to explore
Blanchard, Julien. "Un système de visualisation pour l'extraction, l'évaluation, et l'exploration interactives des règles d'association." Phd thesis, Université de Nantes, 2005. http://tel.archives-ouvertes.fr/tel-00421413.
Full textreprésentation de la connaissance en sciences cognitives. En fouille de données, la principale technique à base de règles est l'extraction de règles d'association, qui a donné lieu à de nombreux travaux de recherche.
La limite majeure des algorithmes d'extraction de règles d'association est qu'ils produisent communément de grandes quantités de règles, dont beaucoup se révèlent même sans aucun intérêt pour l'utilisateur. Ceci s'explique par la nature non supervisée de ces algorithmes : ne considérant aucune variable endogène, ils envisagent dans les règles toutes les combinaisons possibles de variables. Dans la pratique, l'utilisateur ne peut pas exploiter les résultats tels quels directement à la sortie des algorithmes. Un post-traitement consistant en une seconde opération de fouille se
révèle indispensable pour valider les volumes de règles et découvrir des connaissances utiles. Cependant, alors que la fouille de données est effectuée automatiquement par des algorithmes combinatoires, la fouille de règles est une
tâche laborieuse à la charge de l'utilisateur.
La thèse développe deux approches pour assister l'utilisateur dans le post-traitement des règles d'association :
– la mesure de la qualité des règles par des indices numériques,
– la supervision du post-traitement par une visualisation interactive.
Pour ce qui concerne la première approche, nous formalisons la notion d'indice de qualité de règles et réalisons une classification inédite des nombreux indices de la littérature, permettant d'aider l'utilisateur à choisir les indices pertinents pour son besoin. Nous présentons également trois nouveaux indices aux propriétés originales : l'indice
probabiliste d'écart à l'équilibre, l'intensité d'implication entropique, et le taux informationnel. Pour ce qui concerne la seconde approche, nous proposons une méthodologie de visualisation pour l'exploration interactive des règles. Elle
est conçue pour faciliter la tâche de l'utilisateur confronté à de grands ensembles de règles en prenant en compte ses capacités de traitement de l'information. Dans cette méthodologie, l'utilisateur dirige la découverte de connaissances
par des opérateurs de navigation adaptés en visualisant des ensembles successifs de règles décrits par des indices de qualité.
Les deux approches sont intégrées au sein de l'outil de visualisation ARVis (Association Rule Visualization) pour l'exploration interactive des règles d'association. ARVis implémente notre méthodologie au moyen d'une représentation
3D, inédite en visualisation de règles, mettant en valeur les indices de qualité. De plus, ARVis repose sur un algorithme spécifique d'extraction sous contraintes permettant de générer les règles interactivement au fur et à mesure de la navigation de l'utilisateur. Ainsi, en explorant les règles, l'utilisateur dirige à la fois l'extraction et le
post-traitement des connaissances.
Nguyen, Sao Mai. "Un robot curieux pour l’apprentissage actif par babillage d’objectifs : choisir de manière stratégique quoi, comment, quand et de qui apprendre." Thesis, Bordeaux 1, 2013. http://www.theses.fr/2013BOR15223/document.
Full textThe challenges posed by robots operating in human environments on a daily basis and in the long-termpoint out the importance of adaptivity to changes which can be unforeseen at design time. The robot mustlearn continuously in an open-ended, non-stationary and high dimensional space. It must be able to knowwhich parts to sample and what kind of skills are interesting to learn. One way is to decide what to exploreby oneself. Another way is to refer to a mentor. We name these two ways of collecting data sampling modes.The first sampling mode correspond to algorithms developed in the literature in order to autonomously drivethe robot in interesting parts of the environment or useful kinds of skills. Such algorithms are called artificialcuriosity or intrinsic motivation algorithms. The second sampling mode correspond to social guidance orimitation where the teacher indicates where to explore as well as where not to explore. Starting fromthe study of the relationships between these two concurrent methods, we ended up building an algorithmicarchitecture with a hierarchical learning structure, called Socially Guided Intrinsic Motivation (SGIM).We have built an intrinsically motivated active learner which learns how its actions can produce variedconsequences or outcomes. It actively learns online by sampling data which it chooses by using severalsampling modes. On the meta-level, it actively learns which data collection strategy is most efficient forimproving its competence and generalising from its experience to a wide variety of outcomes. The interactivelearner thus learns multiple tasks in a structured manner, discovering by itself developmental sequences
Romat, Hugo. "From data exploration to presentation : designing new systems and interaction techniques to enhance the sense-making process." Thesis, Université Paris-Saclay (ComUE), 2019. http://www.theses.fr/2019SACLS335/document.
Full textDuring the last decade, the amount of data has been constantly increasing. These data can come from several sources such as smartphones, audio recorders, cameras, sensors, simulations, and can have various structure. While computers can help us process these data, human judgment and domain expertise is what turns the data into actual knowledge. However, making sense of this increasing amount of diverse data requires visualization and interaction techniques. This thesis contributes such techniques to facilitate data exploration and presentation, during sense-making activities. In the first part of this thesis, we focus on interactive systems and interaction techniques to support sense-making activities. We investigate how users work with diverse content in order to make them able to externalize thoughts through digital annotations. We present our approach with two systems. The first system, ActiveInk enables the natural use of pen for active reading during a data exploration process. Through a qualitative study with eight participants, we contribute observations of active reading behaviors during data exploration and design principles to support sense-making. The second system, SpaceInk, is a design space of pen & touch techniques that make space for in-context annotations during active reading by dynamically reflowing documents. In the second part, we focus on techniques to visually represent insights and answers to questions that arise during sense-making activities. We focus on one of the most elaborate data structures: multivariate networks, that we visualize using a node-link diagram visualization. We investigate how to enable a flexible iterative design process when authoring node-link diagrams for multivariate networks. We first present a system, Graphies, that enables the creation of expressive node-link diagram visualizations by providing designers with a flexible workflow that streamlines the creative process, and effectively supports quick design iterations. Moving beyond the use of static visual variables in node-link diagrams, we investigated the use of motion to encode data attributes. To conclude, we show in this thesis that the sense-making process can be enhanced in both processes of exploration and presentation, by using ink as a new medium to transition between exploration and externalization, and by following a flexible, iterative process to create expressive data representations. The resulting systems establish a research framework where presentation and exploration are a core part of visual data systems
Lehn, Rémi. "Un système interactif de visualisation et de fouille de règles pour l'extraction de connaissances dans les bases de données." Nantes, 2000. http://www.theses.fr/2000NANT2110.
Full textLavergne, Julien. "Algorithme de fourmis artificielles pour la construction incrémentale et la visualisation interactive de grands graphes de voisinage." Thesis, Tours, 2008. http://www.theses.fr/2008TOUR4049.
Full textWe present in this work a new incremental algorithm for building proximity graphs for large data sets in order to solve a clustering problem. It is inspired from the self-assembly behavior observed in real ants where ants progressively become attached to an existing support and then successively to other attached ants. Each artificial ant represents one data. The way ants move and build a graph depends on the similarity between the data. A graph, built with our method, is well suitable for visualization and interactively exploration depending on the needs of the domain expert. He can visualize the global shape of the graph and locally explore the neighborhood relations with a content-based navigation. Finally, we present different applications of our work as the interactive clustering, the automatic graph construction of documents and an immersion in a virtual reality environment for discovering knowledge in data
Francisci, Dominique. "Techniques d'optimisation pour la fouille de données." Phd thesis, Université de Nice Sophia-Antipolis, 2004. http://tel.archives-ouvertes.fr/tel-00216131.
Full textHeulot, Nicolas. "Etude des projections de données comme support interactif de l’analyse visuelle de la structure de données de grande dimension." Thesis, Paris 11, 2014. http://www.theses.fr/2014PA112127.
Full textThe cost of data acquisition and processing has radically decreased in both material and time. But we also need to analyze and interpret the large amounts of complex data that are stored. Dimensionality is one aspect of their intrinsic complexity. Visualization is essential during the analysis process to help interpreting and understanding these data. Projection represents data as a 2D scatterplot, regardless the amount of dimensions. However, this visualization technique suffers from artifacts due to the dimensionality reduction. Its lack of reliability implies issues of interpretation and trust. Few studies have been devoted to the consideration of the impact of these artifacts, and especially to give feedbacks on how non-expert users can visually analyze projections. The main approach of this thesis relies on an taking these artifacts into account using interactive techniques, in order to allow data scientists or non-expert users to perform a trustworthy visual analysis of projections. The interactive visualization of the proximities applies a coloring of the original proximities relatives to a reference in the data-space. This interactive technique allows revealing projection artifacts in order to help grasping details of the underlying data-structure. In this thesis, we redesign this technique and we demonstrate its potential by presenting two controlled experiments studying the impact of artifacts on the visual analysis of projections. We also present a design-space based on the lens metaphor, in order to improve this technique and to locally visualize a projection free of artifacts issues
Cram, Damien. "Découverte interactive et complète de chroniques : application à la co-construction de connaissances à partir de traces." Thesis, Lyon 1, 2010. http://www.theses.fr/2010LYO10170.
Full textThis thesis deals with the engineering of knowledge dynamics and it focuses on the interactive discovery of knowledge from activity traces. The applicative context targeted by this work is the management of the dynamic aspect of knowledge in Knowledge Management Systems (KMS). Two theoretical contributions are presented in this thesis. Firstly, we propose an iterative and interactive process for the co-construction of dynamic knowledge that requires a dialogue and a cooperation of the machine and humans. Secondly, we present an algorithm for the complete discovery of temporal patterns in sequences of events. This algorithm implements the machine proactive behaviour in this process. Interaction traces are information that users leave when they interact with their environment. This information about users' activities is collected, sometimes intentionally, by the designer of the environment. Interaction traces are represented in an expressive format designed especially for the engineering of interaction traces: the format of modelled traces. Such interaction traces are managed separately in a Trace-Based System (TBS), which can store modelled traces and provides primitive functions to access them. We argue that such interaction traces are potential containers of contextual knowledge about how users behave in their activities mediated by the traced environment. For this reason, interaction traces can be used for building systems that provide contextual assistance to users. We propose an iterative and interactive process for the co-construction of knowledge from traces. In this process, the machine analyses the traces and suggests some behaviour patterns to the human involved in the process. The human validates these patterns if he finds them relevant. If it is not the case, the human elaborates new requests and the machine suggests new candidate patterns, and so on. The idea behind this process was to build a bottom-up knowledge construction approach that takes into account the dynamic and contextual aspects of knowledge. The proactive participation of the machine to this co-construction process implies/requires the development of an algorithm that can extract temporal pattern from interaction traces, that is complete, and that can provide patterns to the human in real time, so that the knowledge co-construction process takes the form of a dialogue between the human and the machine. Chronicles are patterns that can occur in interaction traces and that contain temporal constraints with numerical bounds. The frequent chronicle mining approach we present in this thesis has been designed to implement the machine's behaviour in this process. This algorithm is the first algorithm for chronicle extraction from a sequence of events that is complete. It allows real time interactivity with its users by returning the partial result set of frequent chronicles, at any time. The algorithm supports temporal and structural user constraints pushing, which allows the human to make the chronicle exploration procedure converge more quickly towards the most interesting chronicles. The algorithm can be configured in a way that makes it return the same non-complete chronicle result set as other existing algorithms in the literature. It can also be configured so as to return the complete frequent chronicle set, or to return the complete set of frequent hybrid episodes. Hybrid episodes are summarized forms of chronicles, with a simpler pattern structure that is easier to understand by humans. When compared to existing chronicle mining algorithms with the same conditions, our algorithm shows equivalent time performances. The main inconvenient of the chronicle discovery problem is that the size of the exploration space depends exponentially on the chronicle length. As a result, it is possible to discover only small chronicles in one shot, which implies the need for an iterative and incremental discovery approach [etc...]
Mongy, Sylvain. "Modélisation et analyse du comportement des utilisateurs exploitant des données vidéo." Phd thesis, Université des Sciences et Technologie de Lille - Lille I, 2008. http://tel.archives-ouvertes.fr/tel-00842718.
Full textMongy, Sylvain. "Modélisation et analyse du comportement des utilisateurs exploitant des données vidéo." Electronic Thesis or Diss., Lille 1, 2008. http://www.theses.fr/2008LIL10073.
Full textOur work proposes to analyze users' behavior using video data. Our objective is to contribute understanding how and why users view each video sequence. We present an approach combining intra-video and inter-video behavior analysis. The intra-video level represents the viewing of a video sequence. The inter-video level represents the sessions (linkage between videos viewed by users). An intra-video behavior is defined by a Markov model built using the actions performed during viewings. We cluster these behavior with a new method derived from K-Means adapted to the use of Models (K-Models). We then characteriz several typical behaviors that allows to estimate the level of interest of each video. An inter-video behavior is defined by a session. This session is an ordered sequence of viewings performed by the users. ln order to cluster these sessions, we propose a hierarchical technique, representing clusters by a set of common subsequences enriched by intra-video behaviors. Results from test sets allow to identify observed behaviors and to conclude on the interest of the videos. We also propose a framework on how to integrate our approach in a search engine in order to detect indexing errors and to propose altemate searches to the users
Cordeil, Maxime. "Exploration et exploitation de l’espace de conception des transitions animées en visualisation d’information." Thesis, Toulouse, ISAE, 2013. http://www.theses.fr/2013ESAE0044/document.
Full textData visualizations allow information to be transmitted to users. In order to explore and understand the data, it is often necessary for users to manipulate the display of this data. When manipulating the visualization, visual transitions are necessary to avoid abrupt changes in this visualization, and to allow the user to focus on the graphical object of interest. These visual transitions can be coded as an animation, or techniques that link the data across several displays. The first aim of this thesis was to examine the benefits and properties of animated transitions used to explore and understand large quantities of multidimensional data. In order to do so, we created a taxonomy of existing animated transitions. This taxonomy allowed us to identify that no animated transition currently exists that allows the user to control the direction of objects during the transition. We therefore proposed an animated transition that allows the user to have this control during the animation. In addition, we studied an animated transition technique that uses 3D rotation to transition between visualizations. We identified the advantages of this technique and propose an improvement to the current design. The second objective was to study the visual transitions used in the Air Traffic Control domain. Air Traffic Controllers use a number of visualizations to view vast information which is duplicated in several places: the Radar screen, the strip board, airplane lists (departures/arrivals) etc. Air traffic controllers perform visual transitions as they search between these different displays of information. We studied the way animations can be used in the Air Traffic Control domain by implementing a radar image prototype which combines three visualizations typically used by Air Traffic Controllers
Jacquemont, Stéphanie. "Contributions de l'inférence grammaticale à la fouille de données séquentielles." Phd thesis, Université Jean Monnet - Saint-Etienne, 2008. http://tel.archives-ouvertes.fr/tel-00366358.
Full textDans ce contexte, nous avons montré que l'exploitation brute, non seulement des séquences d'origine mais aussi des automates probabilistes inférés à partir de celles-ci, ne garantit pas forcément une extraction de connaissance pertinente. Nous avons apporté dans cette thèse plusieurs contributions, sous la forme de bornes minimales et de contraintes statistiques, permettant ainsi d'assurer une exploitation fructueuse des séquences et des automates probabilistes. De plus, grâce à notre modèle nous apportons une solution efficace à certaines applications mettant en jeux des problèmes de préservation de vie privée des individus.
Nguyen, Sao Mai. "A Curious Robot Learner for Interactive Goal-Babbling : Strategically Choosing What, How, When and from Whom to Learn." Phd thesis, Université Sciences et Technologies - Bordeaux I, 2013. http://tel.archives-ouvertes.fr/tel-00936992.
Full textBrisson, Laurent. "Intégration de connaissances expertes dans le processus de fouille de données pour l'extraction d'informations pertinentes." Phd thesis, Université Nice Sophia Antipolis, 2006. http://tel.archives-ouvertes.fr/tel-00211946.
Full textFiot, Céline. "Extraction de séquences fréquentes : des données numériques aux valeurs manquantes." Phd thesis, Université Montpellier II - Sciences et Techniques du Languedoc, 2007. http://tel.archives-ouvertes.fr/tel-00179506.
Full textFiot, Céline. "Extraction de séquences fréquentes : des données numériques aux valeurs manquantes." Phd thesis, Montpellier 2, 2007. http://www.theses.fr/2007MON20056.
Full textDe, Runz Cyril. "Imperfection, temps et espace : modélisation, analyse et visualisation dans un SIG archéologique." Phd thesis, Université de Reims - Champagne Ardenne, 2008. http://tel.archives-ouvertes.fr/tel-00560668.
Full textDel, Razo Lopez Federico. "Recherche de sous-structures arborescentes ordonnées fréquentes au sein de bases de données semi-structurées." Phd thesis, Montpellier 2, 2007. http://www.theses.fr/2007MON20040.
Full textL'objectif de cette thèse est de proposer une méthode d'extraction d'arborescences fréquentes. Cette approche est basée sur une représentation compacte des arborescences cherchant à diminuer la consommation de mémoire dans le processus de fouille. En particulier, nous présentons une nouvelle technique de génération d'arborescences candidates visant à réduire leur nombre. Par ailleurs, nous proposons différents algorithmes pour valider le support des arborescences candidates dans une base de données selon divers types de contraintes d'inclusion d'arbres : induite, incrustée et floue. Finalement nous appliquons nos algorithmes à des jeux de données synthétiques et réels et nous présentons les résultats obtenus.
Candillier, Christophe. "Méthodes d'Extraction de Connaissances à partir de Données (ECD) appliquées aux Systèmes d'Information Géographiques (SIG)." Phd thesis, Université de Nantes, 2006. http://tel.archives-ouvertes.fr/tel-00101491.
Full textDel, Razo Lopez Federico. "Recherche de sous-structures arborescentes ordonnées fréquentes au sein de bases de données semi-structurées." Phd thesis, Université Montpellier II - Sciences et Techniques du Languedoc, 2007. http://tel.archives-ouvertes.fr/tel-00203608.
Full textL'objectif de cette thèse est de proposer une méthode d'extraction d'arborescences fréquentes. Cette approche est basée sur une représentation compacte des arborescences cherchant à diminuer la consommation de mémoire dans le processus de fouille. En particulier, nous présentons une nouvelle technique de génération d'arborescences candidates visant à réduire leur nombre. Par ailleurs, nous proposons différents algorithmes pour valider le support des arborescences candidates dans une base de données selon divers types de contraintes d'inclusion d'arbres : induite, incrustée et floue. Finalement nous appliquons nos algorithmes à des jeux de données synthétiques et réels et nous présentons les résultats obtenus.
Coulet, Adrien. "Construction et utilisation d'une base de connaissances pharmacogénomique pour l'intégration de données et la découverte de connaissances." Phd thesis, Université Henri Poincaré - Nancy I, 2008. http://tel.archives-ouvertes.fr/tel-00332407.
Full textCadot, Martine. "Extraire et valider les relations complexes en sciences humaines : statistiques, motifs et règles d'association." Phd thesis, Université de Franche-Comté, 2006. http://tel.archives-ouvertes.fr/tel-00594174.
Full textJanssoone, Thomas. "Analyse de signaux sociaux multimodaux : application à la synthèse d’attitudes sociales chez un agent conversationnel animé." Electronic Thesis or Diss., Sorbonne université, 2018. http://www.theses.fr/2018SORUS607.
Full textDuring an interaction, non-verbal behavior reflects the emotional state of the speaker, such as attitude or personality. Modulations in social signals tell about someone's affective state like variations in head movements, facial expressions or prosody. Nowadays, machines can use embodied conversational agents to express the same kind of social cues. Thus, these agents can improve the quality of life in our modern societies if they provide natural interactions with users. Indeed, the virtual agent must express different attitudes according to its purpose, such as dominance for a tutor or kindness for a companion. Literature in sociology and psychology underlines the importance of the dynamic of social signals for the expression of different affective states. Thus, this thesis proposes models focused on temporality to express a desired affective phenomenon. They are designed to handle social signals that are automatically extracted from a corpus. The purpose of this analysis is the generation of embodied conversational agents expressing a specific stance. A survey of existing databases lead to the design of a corpus composed of presidential addresses. The high definition videos allow algorithms to automatically evaluate the social signals. After a corrective process of the extracted social signals, an agent clones the human's behavior during the addresses. This provides an evaluation of the perception of attitudes with a human or a virtual agent as a protagonist. The SMART model use sequence mining to find temporal association rules in interaction data. It finds accurate temporal information in the use of social signals and links it with a social attitude. The structure of these rules allows an easy transposition of this information to synthesize the behavior of a virtual agent. Perceptual studies validate this approach. A second model, SSN, designed during an international collaboration, is based on deep learning and domain separation. It allows multi-task learning of several affective phenomena and proposes a method to analyse the dynamics of the signals used. These different contributions underline the importance of temporality for the synthesis of virtual agents to improve the expression of certain affective phenomena. Perspectives give recommendation to integrate this information into multimodal solutions
Koptelov, Maksim. "Link prediction in bipartite multi-layer networks, with an application to drug-target interaction prediction." Thesis, Normandie, 2020. http://www.theses.fr/2020NORMC211.
Full textMany aspects from real life with bi-relational structure can be modeled as bipartite networks. This modeling allows the use of some standard solutions for prediction and/or recommendation of new relations between these objects in such networks. Known as the link prediction task, it is a widely studied problem in network science for single graphs, networks assuming one type of interaction between vertices. For multi-layer networks, allowing more than one type of edges between vertices, the problem is not yet fully solved.The motivation of this thesis comes from the importance of an application task, drug-target interaction prediction. Searching valid drug candidates for a given biological target is an essential part of modern drug development. In this thesis, the problem is modeled as link prediction in a bipartite multi-layer network. Modeling the problem in this setting helps to aggregate different sources of information into one single structure and as a result to improve the quality of link prediction.The thesis mostly focuses on the problem of link prediction in bipartite multi-layer networks and makes two main contributions on this topic.The first contribution provides a solution for solving link prediction in the given setting without limiting the number and type of networks, the main constrains of the state of the art methods. Modeling random walk in the fashion of PageRank, the algorithm that we developed is able to predict new interactions in the network constructed from different sources of information. The second contribution, which solves link prediction using community information, is less straight-forward and more dependent on fixing the parameters, but provides better results. Adopting existing community measures for link prediction to the case of bipartite multi-layer networks and proposing alternative ways for exploiting communities, the method offers better performance and efficiency. Additional evaluation on the data of a different origin than drug-target interactions demonstrate the genericness of proposed approach.In addition to the developed approaches, we propose a framework for validation of predicted interactions founded on an external resource. Based on a collection of biomedical concepts used as a knowledge source, the framework is able to perform validation of drug-target pairs using proposed confidence scores. An evaluation of predicted interactions performed on unseen data shows effectiveness of this framework.At the end, a problem of identification and characterization of promiscuous compounds existing in the drug development process is discussed. The problem is solved as a machine learning classification task. The contribution includes graph mining and sampling approaches. In addition, a graphical interface was developed to provide feedback of the result for experts
Leblay, Joffrey. "Vers une nouvelle forme d'accompagnement des processus dans les systèmes interactifs : apport de la fouille de processus et de la recommandation." Thesis, La Rochelle, 2019. http://www.theses.fr/2019LAROS021.
Full textAn information system is a socio-technical system comprising Business Processes and related data. With the development and democratization of IT tools, stored information is getting more important and distributed. The same is true for processes that are becoming increasingly complex and sensitive for organizations. In order to obtain a service, we had to compose business processes to collect information, transform it and reinject it. The objective of this thesis is to explore the problematic of process control in order to give options for the fabrication of a companion that would guide the user when discovering a process. We focused on computer science aspects. In particular, we are studying the possibility of defining a recommendation methodology based on extracted processes and implementing the corresponding software architecture. The works presented are at the interface between several domains. We chose a research approach based on an iterative cycle. After analyzing the field of process mining and recommendation, we concluded that we needed to strengthen our approach to information gathering. This led us to carry out studies on trace-based systems. We then sought to validate the continuity of our approach on a simple case study. It is about personalizing the course of a student during his training. We have set up a demonstrator which, based on the collection of information from previous promotions, extracts knowledge about the students' courses and makes recommendations on the consequences of the course for a particular student. This study allowed us to set up our end-to-end recommendation process and to propose a first sketch of our architecture. We then looked for a more ambitious case study for which no business process was predefined by an expert. We wanted to see if it is possible to identify behaviors and / or strategies of users using a system. We have placed ourselves in a learning context where the learner is involved in a simulation of a micro-world. This case study allowed us to show how to adapt our methodology and how to take contextual data into account. This case study gave rise to an experiment where two groups used our simulator. The first without recommendation, which allowed us to build a set of execution traces that were used to extract the necessary knowledge on our business processes. The second group benefited from our recommendation system. We observed that in the latter group the performance criterion was improved because the trial / error phenomena are considerably reduced. The experience gained during this thesis pushes us to direct our work toward helping to personalize learning trajectory. In particular, with the definition of a class, taking into account the learner's profile, both in terms of knowledge acquired and learning strategies, leads to the creation of a learning path, and therefore a selection of training blocks, which must be personalized. The methodology we have proposed is a brick to build such an ecosystem
Nguyen, Sao Mai. "Un robot curieux pour l'apprentissage actif par babillage d'objectifs : choisir de manière stratégique quoi, comment, quand et de qui apprendre." Phd thesis, Université Sciences et Technologies - Bordeaux I, 2013. http://tel.archives-ouvertes.fr/tel-00977385.
Full textMuhammad, Fuad Muhammad Marwan. "Similarity Search in High-dimensional Spaces with Applications to Time Series Data Mining and Information Retrieval." Phd thesis, Université de Bretagne Sud, 2011. http://tel.archives-ouvertes.fr/tel-00619953.
Full text