Dans cette thèse, nous nous intéressons aux problématiques liées à la classification des formes 3D de corps humains. L'objectif de ce travail de thèse est de définir une méthodologie exhaustive pour obtenir des classes morphologiques caractérisant des formes humaines représentatives d'une population donnée et d'en extraire les morphotypes les plus significatifs. Une analyse géométrique globale de la forme du torse humain par des techniques de classification non-supervisée est mise en œuvre sur une base de données de 478 scans de femmes issues de la campagne de mensuration française réalisée en 2006 par l'IFTH. La description des scans 3D est réalisée avec un calcul des distributions des distances géodésiques à partir d'un ensemble de points anthropométriques positionnés sur la surface du torse 3D. Ce descripteur permet une comparaison quantitative des morphologies dans un monde tridimensionnel. Ces distributions géodésiques sont ensuite utilisées comme entrées pour les deux méthodes de classification non-supervisée que nous avons expérimentées. A l'issue de la classification, trois classes morphologiques, représentées par trois morphotypes, sont extraites dans la base de données. A partir de ces morphotypes, nous avons mis en place une méthode de conception des mannequins adaptatifs. Deux types d'ajustement du mannequin, en fonction de son volume et de sa stature, ont été développés afin de le contrôler à l’aide des individus inclus dans les classes obtenues ou à partir d'un système de taille industriel.Compte tenu de la diversité des disciplines telles que la morphologie humaine, les descripteurs de forme 3D, la classification non-supervisée et les mannequins morphotypes adaptatifs, cette thèse peut contribuer à de nombreuses recherches et problématiques industrielles du secteur de l'habillement. / In this PhD we investigate the problem of 3D human morphologies classification. The aim of this research is to define an exhaustive methodology to obtain a clustering of human morphology shapes representative of a population and to extract the most significant morphotype of each class. An overall geometric shape analysis of body surfaces coupled with technique of unsupervised clustering is implemented on a database of 478 scans of woman bodies from the French Sizing Survey (2006) conducted by IFTH.The description of the 3D scans is performed with a computation of the geodesic distributions based on anthropometrics feature points placed on the surface of human torso. This descriptor enables a quantitative comparison of the morphologies in 3D space. These geodesic distributions are then used as inputs for the two clustering methods we have achieved. After the classification, three morphological classes represented by three morphotypes are extracted from the database. Based on these morphotypes, we developed a design method of adaptive models. Two types of the mannequin adjustment, depending on its volume and height, have been developed to control it via included individuals in the obtained classes or via an industrial size system.

This thesis proposes a new, highly generalised and fundamental, information-modelling framework called the TRM (Ternary Relations Model). The TRM was designed to be a model for converging a number of differing paradigms of information management, some of which are quite isolated. These include areas such as: hypertext navigation; relational databases; semi-structured databases; the Semantic Web; ZigZag and workflow modelling. While many related works model linking by the connection of two ends, the TRM adds a third element to this, thereby enriching the links with associative meanings. The TRM is a formal description of a technique that establishes bi-directional and dynamic node-link structures in which each link is an ordered triple of three other nodes. The key features that makes the TRM distinct from other triple-based models (such as RDF) is the integration of bi-directionality, functional links and simplicity in the definition and elements hierarchy. There are two useful applications of the TRM. Firstly it may be used as a tool for the analysis of information models, to elucidate connections and parallels. Secondly, it may be used as a “construction kit” to build new paradigms and/or applications in information management. The TRM may be used to provide a substrate for building diverse systems, such as adaptive hypertext, schemaless database, query languages, hyperlink models and workflow management systems. It is, however, highly generalised and is by no means limited to these purposes.

La composition monomérique de polymères joue un rôle essentiel dans la comparaison de structures et dans la biologie de synthèse. Cependant, la plupart des ressources moléculaires en ligne donne accès à la structure atomique des molécules et non à leur structure monomérique. Nous avons donc créé un logiciel appelé smiles2monomers (s2m) pour inférer la structure monomérique passer des atomes aux monomères. L’algorithme sous-jacent se déroule en deux phases : une phase de recherche par isomorphisme de sous graphe des monomères au sein de la structure atomique puis une recherche du meilleur pavage non chevauchant des monomères trouvés. La recherche est basée sur un index markovien améliorant les vitesses de recherche de 30% par rapport à l’état de l’art. Le pavage est lui constitué d’un algorithme glouton couplé à un raffinement par “branch & cut”. s2m a été testé sur deux jeux de données déjà annotés. Il retrouve les annotations manuelles avec une excellente sensibilité en des temps très courts. Notre équipe développe Norine, base de données de référence de polymères particuliers appelés Peptides Non Ribosomiques (NRP). s2m, exécuté sur l’ensemble des données de Norine, a mis à jour de nombreuses annotations erronées en base. s2m est donc à la fois capable de créer de nouvelles annotations et d’en corriger des anciennes. Les nouvelles annotations nous servent à la fois à découvrir de nouveaux NRP, de nouvelles fonctionnalités NRP et potentiellement dans le futur à synthétiser des NRP non naturels. / The monomeric composition of polymers is powerful for structure comparison and synthetic biology, among others. However, most of the online molecular resources only provide atomic structures but not monomeric structures. So, we designed a software called smiles2monomers (s2m) to infer monomeric structures from chemical ones. The underlying algorithm is composed of two steps: a search of the monomers using a subgraph isomorphism algorithm fitted to our data and a tiling algorithm to obtain the best coverage of the polymer by non-overlapping monomers. The search is based on a Markovian index improving the execution time by 30% compared to the state of art. The tiling is performed using a greedy algorithm refined by a “branch & cut” algorithm. s2m had been tested on two different already annotated datasets. The software reconstructed the manual annotations with an excellent sensibility in a very short time. Norine database, the reference knowledge base about specific polymers called Non Ri bosomal Peptides (NRP), is developed by our research group. s2m, executed on the Norine database, alerted us about wrong manual annotations. So, s2m not only creates new annotations, but also facilitates the process of annotation curation. The new annotations generated by the software are currently used for the discovery of new NRP, new activities and may be used to create completely new and artificial NRP.

This thesis is concerned with the need to deal with data anomalies, inconsistencies and redundancies within the context of data integration in grids. A data Mapping, Matching and Loading (MML) process that is based on the Grid Staging Catalogue Service (MML-GSCATS) method is identified. In particular, the MML-GSCATS method consists of the development of two mathematical algorithms for the MML processes. Specifically it defines an intermediate data storage staging facility in order to process, upload and integrate data from various small to large size data repositories. With this in mind, it expands the integration notion of a database management system (DBMS) to include the MML-GSCATS method in traditional distributed and grid environments. The data mapping employed is in the form of value correspondences between source and target databases whilst data matching consolidates distinct catalogue schemas of federated databases to access information seamlessly. There is a need to deal with anomalies and inconsistencies in the grid, MML processes are applied using a healthcare case study with developed scenarios. These scenarios were used to test the MML-GSCATS method with the help of software prototyping toolkit. Testing has set benchmarks, performance, reliability and error detections (anomalies and redundancies). Cross-scenario data sets were formulated and results of scenarios were compared with benchmarking. These benchmarks help in comparing the MMLGSCATS methodology with traditional and current grid methods. Results from the testing and experiments demonstrate that the MML-GSCATS is a valid method for identifying data anomalies, inconsistencies and redundancies that are produced during loading. Testing results indicates the MML-GSCATS is better than traditional methods.

Μέσα από αυτή την εργασία προσεγγίσαμε την έννοια των οντολογιών και εξετάσαμε μερικά είδη Οντολογιών καθώς και μια υφιστάμενη κατηγοριοποίηση. Αναφέραμε γλώσσες αναπαράστασης Οντολογιών, και αναλύσαμε τις σημαντικότερες από αυτές. Περιγράψαμε συνοπτικά μερικά εργαλεία κατασκευής οντολογιών με μεγαλύτερη έμφαση στο λογισμικό PROTÉGÉ 2000. Θίξαμε θέματα που αφορούν την μηχανική οντολογιών και παρουσιάσαμε περιληπτικά μερικές μεθοδολογίες ανάπτυξης Οντολογιών. Αναλύσαμε σε θεωρητικό επίπεδο τα βασικά βήματα ανάπτυξης μιας οντολογίας μέσα από ένα υπαρκτό παράδειγμα και παρουσιάσαμε την ανάπτυξη μιας οντολογίας που πραγματοποιήθηκε σε περιβάλλον του λογισμικού PROTÉGÉ για τις ανάγκες της παρούσας διπλωματικής. / Through this work we approached the concept of Ontologies and discussed an existing classification. We mentioned Ontologies representation languages, and we analyzed the most important of these. We outlined some ontologies construction tools with a greater emphasis on software PROTÉGÉ 2000. We raised issues relating to engineering Ontologies and we summarized some methodologies. We analyzed, at the theoretical level, the basic steps in developing an Ontology through a real example and we showed the development of an ontology held in PROTÉGÉ 2000 software environment for the purposes of this project.

This thesis examines the contribution of a guided discovery learning approach within a simulated real world context to learning. In order to consider the potential of this approach, a database design task is chosen (Storey & Goldstein, 1993) which requires the learner to capture the semantics of the domain application in a real world situation and then translate this into a data model for the database management system. This approach to learning has advantages since simulating a real world system in a classroom can be a very difficult and time-consuming activity. The aims of the thesis is, therefore, to investigate the possibility of simulating real world situation for gathering database requirements and a teaching strategy that is suitable for this real world situation context. In order to reach the research goal, two main research questions need to be answered. Firstly, to what extent can a simulation of a real world situation improve the quality of learning in the database design area? Secondly, the extent to which a guided discovery teaching strategy can enhance the learning of database design within such a (simulated) real world context? A framework for simulating the real world situation and guided discovery strategies had been designed in order to implement four versions of a prototype systems called GERM for evaluation in order to answer the research questions. The main results obtained from a small group of learners and lecturers indicates that the potential of guided discovery learning within a real world context can improve the quality of learning in database design - in particular entity relationship modelling. Amongst other advantages, it can help students to change their basic misconceptions. Furthermore, it also can improve students' skills in a real world situation. The promising results suggest further lines of research.

Many businesses produce documents as part of their daily activities: software engineers produce requirements specifications, design models, source code, build scripts and more; business analysts produce glossaries, use cases, organisation charts, and domain ontology models; service providers and retailers produce catalogues, customer data, purchase orders, invoices and web pages. What these examples have in common is that the content of documents is often semantically related: source code should be consistent with the design model, a domain ontology may refer to employees in an organisation chart, and invoices to customers should be consistent with stored customer data and purchase orders. As businesses grow and documents are added, it becomes difficult to manually track and check the increasingly complex relationships between documents. The problem is compounded by current trends towards distributed working, either over the Internet or over a global corporate network in large organisations. This adds complexity as related information is not only scattered over a number of documents, but the documents themselves are distributed across multiple physical locations. This thesis addresses the problem of managing the consistency of distributed and possibly heterogeneous documents. “Documents” is used here as an abstract term, and does not necessarily refer to a human readable textual representation. We use the word to stand for a file or data source holding structured information, like a database table, or some source of semi-structured information, like a file of comma-separated values or a document represented in a hypertext markup language like XML [Bray et al., 2000]. Document heterogeneity comes into play when data with similar semantics is represented in different ways: for example, a design model may store a class as a rectangle in a diagram whereas a source code file will embed it as a textual string; and an invoice may contain an invoice identifier that is composed of a customer name and date, both of which may be recorded and managed separately. Consistency management in this setting encompasses a number of steps. Firstly, checks must be executed in order to determine the consistency status of documents. Documents are inconsistent if their internal elements hold values that do not meet the properties expected in the application domain or if there are conflicts between the values of elements in multiple documents. The results of a consistency check have to be accumulated and reported back to the user. And finally, the user may choose to change the documents to bring them into a consistent state. The current generation of tools and techniques is not always sufficiently equipped to deal with this problem. Consistency checking is mostly tightly integrated or hardcoded into tools, leading to problems with extensibility with respect to new types of documents. Many tools do not support checks of distributed data, insisting instead on accumulating everything in a centralized repository. This may not always be possible, due to organisational or time constraints, and can represent excessive overhead if the only purpose of integration is to improve data consistency rather than deriving any additional benefit. This thesis investigates the theoretical background and practical support necessary to support consistency management of distributed documents. It makes a number of contributions to the state of the art, and the overall approach is validated in significant case studies that provide evidence of its practicality and usefulness.

Traditional data integration offers high quality services for managing and querying interrelated but heterogeneous data sources but at a high cost. This is because a significant amount of manual effort is required to help specify precise relationships between the data sources in order to set up a data integration system. The recent proposed vision of dataspaces aims to reduce the upfront effort required to set up the system. A possible solution to approaching this aim is to infer schematic correspondences between the data sources, thus enabling the development of automated means for bootstrapping dataspaces. In this thesis, we discuss a two-step research programme to automatically infer schematic correspondences between data sources. In the first step, we investigate the effectiveness of existing schema matching approaches for inferring schematic correspondences and contribute a benchmark, called MatchBench, to achieve this aim. In the second step, we contribute an evolutionary search method to identify the set of entity-level relationships (ELRs) between data sources that qualify as entity-level schematic correspondences. Specifically, we model the requirements using a vector space model. For each resulting ELR we further identify a set of attribute-level relationships (ALRs) that qualify as attribute-level schematic correspondences. We demonstrate the effectiveness of the contributed inference technique using both MatchBench scenarios and real world scenarios.

Les algorithmes d'alignement sont au coeur de l'analyse de séquences en bio-informatique. Dans cette thèse, nous nous focalisons sur le problème de l'alignement de lectures, des millions de courtes séquences produites par les séquenceurs de nouvelle génération (NGS) en particulier pour l'analyse de données de métatranscriptome et de métagénome en biodiversité. Pour cela, il y a deux types de difficulté. Le premier est que toutes les technologies NGS entrainent des erreurs de séquençage, telles que substitutions, insertions et suppressions de nucléotides. Le second est que les échantillons métagénomique peuvent contenir des centaines d'organismes inconnus et que leur analyse demande de procéder à des alignements avec des d'espèces possiblement distantes. Pour résoudre ces problèmes, nous avons développé un nouvel algorithme d'alignement reposant sur des graines avec erreurs. Cela amène un gain en sensibilité par rapport aux logiciels existants optimisés pour le problème du reséquençage, avec des similarités élevées et qui se fondent sur des graines exactes. Nous proposons également une nouvelle méthode d'indexation basée sur le Burst trie qui permet d'optimiser la recherche avec les graines avec erreurs. Nous montrons l'efficacité de nos méthodes dans deux nouveaux outils, SortMeRNA pour l'identification d'ARN ribosomiques dans des données de métatranscriptome, et SortMeDNA pour l'alignement de lectures en génomique et métagénomique. / Sequence alignment algorithms are at the heart of bioinformatic sequence analysis. In this thesis we focus on the alignment of millions of short sequences produced by Next-Generation Sequencing (NGS) technologies in particular for the analysis of metagenomic and metatranscriptomic data, that is the DNA and RNA directly extracted for an environment. Two major challenges were confronted in our developed algorithms. First, all NGS technologies today are susceptible to sequencing errors in the form of nucleotide substitutions, insertions and deletions. Second, metagenomic samples can contain hundreds of unknown organisms and the standard approach to identifying them is to align against known closely related species. To overcome these challenges we designed a new approximate matching technique based on the universal Levenshtein automaton which quickly locates short regions of similarity (seeds) between two sequences allowing 1 error of any type. Using seeds to detect possible high scoring alignments is a widely used heuristic for rapid sequence alignment, although most existing software are optimized for performing high similarity searches and apply exact seeds. Furthermore, we describe a new indexing data structure based on the Burst trie which optimizes the search for approximate seeds. We demonstrate the efficacy of our method in two implemented software, SortMeRNA and SortMeDNA. The former can quickly filter ribosomal RNA fragments from metatranscriptomic data and the latter performs full alignment for genomic and metagenomic data.

We aim to develop a formal framework to reason about provenance in distributed systems. We take as our starting point an extension of the asynchronous pi-calculus where processes are explicitly assigned principal identities. We enrich this basic setting with provenance annotated data, dynamic provenance tracking and dynamically checked trust policies. We give several examples to illustrate the use of the calculus in modelling systems where principals base their trust in the quality of data on the provenance information associated with it. We consider the role of provenance in the calculus by relating the provenance tracking semantics to a plain one in which no provenance tracking or checking takes place. We further substantiate this by studying bisimulation-based behavioural equivalences for the plain and annotated versions of the calculus and contrasting the discriminating power of the equivalences obtained in each case. We also give a more denotational take on the semantics of the provenance calculus and look at notions of well-formedness and soundness for the provenance tracking semantics. We consider two different extensions of the basic calculus. The first aims to alleviate the cost of run time provenance tracking and checking by defining a static type system which guarantees that in well-typed systems principals always receive data with provenance that matches their requirements. The second extension looks at the ramifications of provenance tracking on privacy and security policies and consists of extending the calculus with a notion we call filters. This gives principals the ability to assign different views of the provenance of a given value to different principals, thus allowing for the selective disclosure of provenance information. We study behavioural equivalences for this extension of the calculus, paying particular attention to the set of principals composing the observer and its role in discriminating between systems.