O reconhecimento de expressões faciais é uma área de interesse da ciência da computação e tem sido um atrativo para pesquisadores de diferentes áreas, pois tem potencial para promover o desenvolvimento de diferentes tipos de aplicações. Reconhecer automaticamente essas expressões tem se tornado um objetivo, principalmente na área de análise do comportamento humano. Especialmente para estudo das línguas de sinais, a análise das expressões faciais é importante para a interpretação do discurso, pois é o elemento que permite expressar informação prosódica, suporta o desenvolvimento da estrutura gramatical e semântica da língua, e ajuda na formação de sinais com outros elementos básicos da língua. Nesse contexto, as expressões faciais são chamadas de expressões faciais gramaticais e colaboram na composição no sentido semântico das sentenças. Entre as linhas de estudo que exploram essa temática, está aquela que pretende implementar a análise automática da língua de sinais. Para aplicações com objetivo de interpretar línguas de sinais de forma automatizada, é preciso que tais expressões sejam identificadas no curso de uma sinalização, e essa tarefa dá-se é definida como segmentação de expressões faciais gramaticais. Para essa área, faz-se útil o desenvolvimento de uma arquitetura capaz de realizar a identificação de tais expressões em uma sentença, segmentando-a de acordo com cada tipo diferente de expressão usada em sua construção. Dada a necessidade do desenvolvimento dessa arquitetura, esta pesquisa apresenta: uma análise dos estudos na área para levantar o estado da arte; a implementação de algoritmos de reconhecimento de padrões usando Multilayer Perceptron e misturas de especialistas para a resolução do problema de reconhecimento da expressão facial; a comparação desses algoritmos reconhecedores das expressões faciais gramaticais usadas na concepção de sentenças na Língua Brasileira de Sinais (Libras). A implementação e teste dos algoritmos mostraram que a segmentação automática de expressões faciais gramaticais é viável em contextos dependentes do usuários. Para contextos independentes de usuários, o problema de segmentação de expressões faciais representa um desafio que requer, principalmente, a organização de um ambiente de aprendizado estruturado sobre um conjunto de dados com volume e diversidade maior do que os atualmente disponíveis / The recognition of facial expressions is an area of interest in computer science and has been an attraction for researchers in different fields since it has potential for development of different types of applications. Automatically recognizing these expressions has become a goal primarily in the area of human behavior analysis. Especially for the study of sign languages, the analysis of facial expressions represents an important factor for the interpretation of discourse, since it is the element that allows expressing prosodic information, supports the development of the grammatical and semantic structure of the language, and eliminates ambiguities between similar signs. In this context, facial expressions are called grammatical facial expressions. These expressions collaborate in the semantic composition of the sentences. Among the lines of study that explore this theme is the one that intends to implement the automatic analysis of sign language. For applications aiming to interpret signal languages in an automated way, it is necessary that such expressions be identified in the course of a signaling, and that task is called \"segmentation of grammatical facial expressions\'\'. For this area, it is useful to develop an architecture capable of performing the identification of such expressions in a sentence, segmenting it according to each different type of expression used in its construction. Given the need to develop this architecture, this research presents: a review of studies already carried out in the area; the implementation of pattern recognition algorithms using Multilayer Perceptron and mixtures of experts to solve the facial expression recognition problem; the comparison of these algorithms as recognizers of grammatical facial expressions used in the conception of sentences in the Brazilian Language of Signs (Libras). The implementation and tests carried out with such algorithms showed that the automatic segmentation of grammatical facial expressions is practicable in user-dependent contexts. Regarding user-independent contexts, this is a challenge which demands the organization of a learning environment structured on datasets bigger and more diversified than those current available

Made available in DSpace on 2016-08-10T10:46:45Z (GMT). No. of bitstreams: 1 MURILO TEIXEIRA COSTA.pdf: 719987 bytes, checksum: eca9173d0b9f6dbaf05c3cf0233259e7 (MD5) Previous issue date: 2012-03-16 / It is intended to develop in this dissertation research on the limits and possibilities of public hearings with experts in the abstract control of constitutionality. Public hearings may be defined as a participatory mechanism provided for in the process of abstract judicial review, in an advisory and non-binding, which create opportunities to direct contact between judge and expert, are intended to hear the varied opinions on certain submission to trial, supporting thus the performance of judicial functions. This dissertation intends to specifically investigate whether the opinions and knowledge exhibited at such events has, in fact, interfered in any way in the activity of constitutional interpretation and judgments of the Supreme Court. A question that should properly be made on the effectiveness of such hearings refers to the fact whether they really have contributed to a pluralistic and democratic opening in brazilian process of constitutional interpretation, or whether, conversely, has been used only to meet a mere ritualism or formalism, without any influence on the process of constitutional interpretation and decisions of the Supreme Court. This study acquires significant importance today when the Supreme Court has been constantly driven to solve extremely difficult and multidisciplinary causes, which has attracted the attention of public opinion and the community, noted that in some of these moments, the Supreme Court triggered the mechanism the public hearing and promoted the hearing of experts as well as other segments of society are directly interested in the affairs ended in actions such as associations and other representative groups. The brazilian legal literature is still lacking specific scientific papers on this subject, although I count on a big production about the institution of amicus curiae. The relevance of the subject of public hearings with experts in the abstract control of constitutionality can not be overlooked, since it interacts primarily with the issue pertaining to the active participation of citizens, groups, community, the public powers and public opinion in the process of constitutional interpretation. A set of ideas with this meaning has a central place in the work of the german jurist Peter Häberle, which advocates the adoption of constitutional hermeneutics criteria for an open society that is increasingly pluralistic. This research focuses on investigating also the public hearings with experts can really contribute to that court decisions are more fair and legitimate. Held, in addition to literature review, a case study of public hearing in the Supreme Court in 2007, ADI 3510 which discussed the issue pertaining to permission to use embryonic stem cells for research and therapy, specifically for check whether the opinion and knowledge of experts affect the positioning of the ministers of the Supreme Court in the trial of this action. One hopes, to investigate the dynamic that has been given to such public hearings, build a study of its limits and its possibilities. / Pretende-se nesta dissertação desenvolver uma pesquisa sobre os limites e as possibilidades das audiências públicas realizadas com especialistas no âmbito do controle abstrato de constitucionalidade. As audiências públicas podem ser definidas como um mecanismo participativo previsto no processo de controle abstrato de constitucionalidade, com caráter consultivo e não-vinculante, que, ao oportunizar o contato direto entre julgador e especialista, destinam-se a ouvir as mais variadas opiniões sobre determinada matéria submetida a julgamento, subsidiando, dessa forma, o desempenho da função jurisdicional. Esta dissertação pretende especialmente investigar se as opiniões e os conhecimentos exposto em tais eventos tem, de fato, interferido de alguma forma na atividade de interpretação constitucional e nos julgamentos do STF. Um questionamento que deve ser validamente formulado sobre a eficácia de tais audiências refere-se ao fato de saber se elas, realmente, tem contribuído para uma abertura pluralista e democrática no processo brasileiro de interpretação constitucional, ou se, pelo contrário, tem sido utilizadas apenas para atender a um mero ritualismo ou formalismo, sem qualquer influência no processo de interpretação constitucional e de tomada de decisões do STF. Esse estudo adquire significativa importância nos dias atuais em que o STF tem sido constantemente acionado para solucionar causas extremamente difíceis e multidisciplinares, as quais tem atraído a atenção da opinião pública e da comunidade, observado que, em alguns desses momentos, o STF acionou o mecanismo de realização de audiência pública e promoveu a oitiva de especialistas e também de outros segmentos da sociedade diretamente interessados nos assuntos encerrados nas ações, como associações e outros grupos de representação. A literatura jurídica brasileira ainda é carente de trabalhos científicos específicos sobre este tema, embora já conte com uma grande produção sobre o instituto do amicus curiae. A relevância da temática das audiências públicas com especialistas no âmbito do controle abstrato de constitucionalidade não pode passar despercebido, posto que interage fundamentalmente com a questão pertinente à participação ativa dos cidadãos, dos grupos, da comunidade, das potências públicas e da opinião pública no processo de interpretação constitucional. Um conjunto de idéias com esse significado tem lugar central na obra do jurista alemão Peter Häberle, o qual defende a adoção de critérios de hermenêutica constitucional abertos para uma sociedade cada vez mais pluralista. Esta pesquisa preocupa-se em investigar também se as audiências públicas com especialistas podem realmente contribuir para que as decisões judiciais sejam mais justas e legítimas. Realiza-se, além da revisão da literatura, um estudo de caso da audiência pública realizada no STF, em 2007, na ADI 3510 que debateu a questão pertinente à permissão do uso de células-tronco embrionárias para fins de pesquisas e terapia, especificamente para conferir se a opinião e o conhecimento dos especialistas influenciaram o posicionamento dos ministros do STF no julgamento desta ação. Almeja-se, ao pesquisar a dinâmica que tem sido conferida a tais audiências públicas, construir um estudo sobre os seus limites e as suas possibilidades.

Mot valise véhiculé par les agences internationales, la notion de « développement avec identité » traduit une volonté de prendre en compte les spécificités locales et culturelles des populations autochtones et la promotion d’une approche en terme de développement, présenté comme universel et apolitique. C’est cette contradiction que cette thèse propose d’explorer. À partir du cas mexicain, il s’agit de comprendre comment les dispositifs de développement fonctionnent comme des instruments de connaissance, mais également comme des instruments performatifs par leurs inductions prescriptives et les usages sociaux qui en sont faits. Au final, le « développement avec identité » renvoie à un « art de gouverner » les populations marquées par une disqualification sociale. Celui-ci s’appuie sur deux volets : l’élaboration de savoirs spécifiques sur ces populations et la participation des individus et groupes autochtones aux dispositifs de développement. Cette thèse propose une contribution sociologique à l’analyse du développement et aux travaux sur la question autochtone. Cette étude de la circulation des discours, des instruments et des pratiques du développement se fonde sur une enquête ethnographique de trois espaces où se construisent, s’élaborent et se mettent en pratique les dispositifs du développement autochtone au Mexique : le bureau du Programme des Nations unies à Mexico, l’instance nationale en charge de l’action publique de développement des populations autochtones, et enfin des groupes sociaux de la région du Chiapas dans lesquels des dispositifs de développement sont également élaborés et mis en œuvre / A buzzword broadcasted by international agencies, the concept of “development with identity” refers to a willingness to take into account local and cultural specificities of indigenous peoples. It entails a promotion of the development approach, presented as both universal and apolitical. This thesis proposes to explore this contradiction. Focusing on the Mexican case, it aims to understand how development apparatuses function as instruments of knowledge, but also as performative instruments by their prescriptive inductions and the social uses they are made of. Finally, “development with identity” refers to an “art of government” of populations labeled as socially disqualified. It relies on two key components: the elaboration of a specific knowledge on these populations and the participation of indigenous individuals and groups in development apparatuses.This thesis proposes a sociological contribution to the analysis of development and works on indigenous issues. In order to study the circulation of development speeches, instruments and practices, this work is based on a political ethnography of three spaces in which the apparatuses of indigenous development in Mexico are elaborated, formulated and put into practice: the Office of the United Nations Development Program in Mexico, the national authority in charge of development public policy for indigenous peoples, and finally social groups in the region of Chiapas in which development apparatuses are developed and implemented

La reconnaissance des perturbations survenant sur les réseaux HTA est une problématique essentielle pour les clients industriels comme pour le gestionnaire du réseau. Ces travaux de thèse ont permis de développer un système d’identification automatique. Il s’appuie sur des méthodes de segmentation qui décomposent de manière précise et efficace les régimes transitoires et permanents des perturbations. Elles utilisent des filtres de types Kalman linéaire ou anti-harmoniques pour extraire les régimes transitoires. La prise en compte des variations harmoniques et de la présence de transitoires proches se fait à l’aide de seuils adaptatifs. Des méthodes de correction du retard a posteriori permettent d’améliorer la précision de la décomposition. Des indicateurs adaptés à la dynamique des régimes de fonctionnement analysés sont utilisés pour caractériser les perturbations. Peu sensibles aux erreurs de segmentation et aux perturbations harmoniques, ils permettent une description fiable des phases des perturbations. Deux types de systèmes de décision ont également été étudiés : des systèmes experts et des classifieurs SVM. Ces systèmes ont été mis au point à partir d’une large base de perturbations simulées. Leurs performances ont été évaluées sur une base de perturbations réelles : ils déterminent efficacement le type et la direction des perturbations observées (taux de reconnaissance moyen > 98%). / The recognition of disturbances affecting MV networks is essential to industrials and distribution system operators. The aim of this thesis work is to design a near real-time automatic system able to detect and identify disturbances from their waveforms. Segmentation methods split the disturbed waveforms into transient and steady-state intervals. They use Kalman filters or anti-harmonic filters to extract the transient intervals. Adaptive thresholding methods increase the detection capacity while a posterior delay compensation methods improve the accuracy of the decomposition. Indicators adapted to the disturbance dynamic are used to characterize its steady-state and transient phases. They are robust to segmentation inaccuracies as well as to steady-state disturbances such as harmonics. Two distinct decision systems are also studied: expert recognition systems and SVM classifiers. During the learning stage, a large simulated event database is used to train both systems. Their performances are evaluated on real events: the type and direction of the measured disturbances are determined with a recognition rate over 98%.

O processo penal no Brasil é conjunto de atos jurídicos, formalmente ordenados, com o objetivo de apurar a veracidade de fatos criminosos, resultando na sentença condenatória ou absolvição do agente envolvido no crime. Ocorre que durante o procedimento inclui a figura do perito criminal como representante legal para a análise de provas das infrações. As diversas modalidades do crime resultam em complexos levantamentos de provas, exigindo do perito criminal o conhecimento técnico - científico do crime específico a ser examinado. Os Institutos de perícias são formados por profissionais multidisciplinares que tem como objetivo atingir a maior abrangência das ciências para auxiliar na justiça. Com a nova modalidade de crime de maus-tratos aos animais, o Estado tem a responsabilidade de materializar os vestígios relacionados com os animais, convergindo o respectivo trabalho ao perito criminal com formação em medicina veterinária. Atualmente é reduzido o número profissionais veterinários nos quadros dos peritos criminais, interferindo nos compromissos da veracidade ou autenticidade dos fatos. Paralelamente a esta nova demanda, os institutos de ensino superior estão implementando a nova especialidade denominada de Medicina Veterinária Legal que tem como objetivo disciplinar o conhecimento da medicina veterinária no auxilio da justiça. O presente trabalho foi dividido em duas partes, onde a primeira tem como objetivo expor o funcionamento sistema jurídico nacional e as atribuições dos peritos criminais, pontuando as importâncias do conhecimento médico veterinário legal quando do levantamento de provas envolvendo maus-tratos aos animais, e na segunda parte, as imagens das práticas do levantamento de provas pelo perito criminal médico veterinário em infrações de maus-tratos aos animais e discussões envolvendo as dificuldades e apontamentos pertinentes a prática legal. / Criminal processes in Brazil are a set of legal acts formally organized with the goal of investigating the veracity of crime facts, resulting in a sentence or pardon of the person involved in the crime. During the process, the forensic expert is included as a legal representative to analyze the evidence. Several modalities of crime result in complex retrieval of evidence, requiring from the forensic expert technical and scientific knowledge of the crime to be examined. Forensic Institutes are composed of multidisciplinary professionals with the goal of achieving the largest possible breadth of sciences to assist the Law. In the new animal abuse modality of crime, the State has the responsibility of retrieving the traces related to animals, turning over the work to the forensic expert with training in veterinary medicine. Currently the number of veterinarians is small among forensic experts, interfering with veracity and authenticity of facts. In parallel to this new demand, higher education institutions are implementing a new specialty called Legal Veterinary Medicine, which aims at formalizing the knowledge of veterinary medicine to assist the Law. The current work is divided into two parts: the first aims at exposing how the national legal system works and the attributions of forensic experts, evidencing the importance of Legal Veterinary Medicine when retrieving evidence involving animal abuse. And the second part exposes the images of the practices of evidence retrieval by forensic veterinary medics in animal abuse cases and discussions involving the difficulties and observations relative to legal practices.

O reconhecimento de expressões faciais é uma área de interesse da ciência da computação e tem sido um atrativo para pesquisadores de diferentes áreas, pois tem potencial para promover o desenvolvimento de diferentes tipos de aplicações. Reconhecer automaticamente essas expressões tem se tornado um objetivo, principalmente na área de análise do comportamento humano. Especialmente para estudo das línguas de sinais, a análise das expressões faciais é importante para a interpretação do discurso, pois é o elemento que permite expressar informação prosódica, suporta o desenvolvimento da estrutura gramatical e semântica da língua, e ajuda na formação de sinais com outros elementos básicos da língua. Nesse contexto, as expressões faciais são chamadas de expressões faciais gramaticais e colaboram na composição no sentido semântico das sentenças. Entre as linhas de estudo que exploram essa temática, está aquela que pretende implementar a análise automática da língua de sinais. Para aplicações com objetivo de interpretar línguas de sinais de forma automatizada, é preciso que tais expressões sejam identificadas no curso de uma sinalização, e essa tarefa dá-se é definida como segmentação de expressões faciais gramaticais. Para essa área, faz-se útil o desenvolvimento de uma arquitetura capaz de realizar a identificação de tais expressões em uma sentença, segmentando-a de acordo com cada tipo diferente de expressão usada em sua construção. Dada a necessidade do desenvolvimento dessa arquitetura, esta pesquisa apresenta: uma análise dos estudos na área para levantar o estado da arte; a implementação de algoritmos de reconhecimento de padrões usando Multilayer Perceptron e misturas de especialistas para a resolução do problema de reconhecimento da expressão facial; a comparação desses algoritmos reconhecedores das expressões faciais gramaticais usadas na concepção de sentenças na Língua Brasileira de Sinais (Libras). A implementação e teste dos algoritmos mostraram que a segmentação automática de expressões faciais gramaticais é viável em contextos dependentes do usuários. Para contextos independentes de usuários, o problema de segmentação de expressões faciais representa um desafio que requer, principalmente, a organização de um ambiente de aprendizado estruturado sobre um conjunto de dados com volume e diversidade maior do que os atualmente disponíveis / The recognition of facial expressions is an area of interest in computer science and has been an attraction for researchers in different fields since it has potential for development of different types of applications. Automatically recognizing these expressions has become a goal primarily in the area of human behavior analysis. Especially for the study of sign languages, the analysis of facial expressions represents an important factor for the interpretation of discourse, since it is the element that allows expressing prosodic information, supports the development of the grammatical and semantic structure of the language, and eliminates ambiguities between similar signs. In this context, facial expressions are called grammatical facial expressions. These expressions collaborate in the semantic composition of the sentences. Among the lines of study that explore this theme is the one that intends to implement the automatic analysis of sign language. For applications aiming to interpret signal languages in an automated way, it is necessary that such expressions be identified in the course of a signaling, and that task is called \"segmentation of grammatical facial expressions\'\'. For this area, it is useful to develop an architecture capable of performing the identification of such expressions in a sentence, segmenting it according to each different type of expression used in its construction. Given the need to develop this architecture, this research presents: a review of studies already carried out in the area; the implementation of pattern recognition algorithms using Multilayer Perceptron and mixtures of experts to solve the facial expression recognition problem; the comparison of these algorithms as recognizers of grammatical facial expressions used in the conception of sentences in the Brazilian Language of Signs (Libras). The implementation and tests carried out with such algorithms showed that the automatic segmentation of grammatical facial expressions is practicable in user-dependent contexts. Regarding user-independent contexts, this is a challenge which demands the organization of a learning environment structured on datasets bigger and more diversified than those current available

O processo penal no Brasil é conjunto de atos jurídicos, formalmente ordenados, com o objetivo de apurar a veracidade de fatos criminosos, resultando na sentença condenatória ou absolvição do agente envolvido no crime. Ocorre que durante o procedimento inclui a figura do perito criminal como representante legal para a análise de provas das infrações. As diversas modalidades do crime resultam em complexos levantamentos de provas, exigindo do perito criminal o conhecimento técnico - científico do crime específico a ser examinado. Os Institutos de perícias são formados por profissionais multidisciplinares que tem como objetivo atingir a maior abrangência das ciências para auxiliar na justiça. Com a nova modalidade de crime de maus-tratos aos animais, o Estado tem a responsabilidade de materializar os vestígios relacionados com os animais, convergindo o respectivo trabalho ao perito criminal com formação em medicina veterinária. Atualmente é reduzido o número profissionais veterinários nos quadros dos peritos criminais, interferindo nos compromissos da veracidade ou autenticidade dos fatos. Paralelamente a esta nova demanda, os institutos de ensino superior estão implementando a nova especialidade denominada de Medicina Veterinária Legal que tem como objetivo disciplinar o conhecimento da medicina veterinária no auxilio da justiça. O presente trabalho foi dividido em duas partes, onde a primeira tem como objetivo expor o funcionamento sistema jurídico nacional e as atribuições dos peritos criminais, pontuando as importâncias do conhecimento médico veterinário legal quando do levantamento de provas envolvendo maus-tratos aos animais, e na segunda parte, as imagens das práticas do levantamento de provas pelo perito criminal médico veterinário em infrações de maus-tratos aos animais e discussões envolvendo as dificuldades e apontamentos pertinentes a prática legal. / Criminal processes in Brazil are a set of legal acts formally organized with the goal of investigating the veracity of crime facts, resulting in a sentence or pardon of the person involved in the crime. During the process, the forensic expert is included as a legal representative to analyze the evidence. Several modalities of crime result in complex retrieval of evidence, requiring from the forensic expert technical and scientific knowledge of the crime to be examined. Forensic Institutes are composed of multidisciplinary professionals with the goal of achieving the largest possible breadth of sciences to assist the Law. In the new animal abuse modality of crime, the State has the responsibility of retrieving the traces related to animals, turning over the work to the forensic expert with training in veterinary medicine. Currently the number of veterinarians is small among forensic experts, interfering with veracity and authenticity of facts. In parallel to this new demand, higher education institutions are implementing a new specialty called Legal Veterinary Medicine, which aims at formalizing the knowledge of veterinary medicine to assist the Law. The current work is divided into two parts: the first aims at exposing how the national legal system works and the attributions of forensic experts, evidencing the importance of Legal Veterinary Medicine when retrieving evidence involving animal abuse. And the second part exposes the images of the practices of evidence retrieval by forensic veterinary medics in animal abuse cases and discussions involving the difficulties and observations relative to legal practices.

Book Summary: Volume 2 of Group Work Experts Share Their Favorite Activities for Supervision is a follow-up to the first volume of this handy resource. This second volume contains more than 50 additionalactivities for the supervision of group work. Activities are organized into four sections: Multicultural and Social Justice Awareness, Supporting Process in Task and PsychoeducationalGroups, Setting and Population Specific interventions, and Developing Skills and the Coleader Relationship in Group Supervision of Group Leaders. Each section begins with adetailed introduction by an expert who reviews the current literature. The multicultural andsocial justice activities identify and address the ways in which identity affects the group as-a-wholeexperience, and also provide a model for supervisor competence. The activities in the Supporting Process in Task and Psychoeducational Groups section are designed to help supervisors more effectively develop and intentionally intervene in these types of groups. The Setting and Population Specific section offers activities for supervisors’ use with unique types of groups across varied contexts, and the development skills and coleader relationship activities support both beginning and more advanced group counselors in their work.

Les logiciels dépendent de bibliothèques tierces pour réduire les coûts liés à leur développement et à leur maintenance. Elles proposent un ensemble de fonctionnalités robustes dont les développeurs peuvent tirer parti depuis une interface de programmation. Cependant, cette forte dépendance entre un logiciel et ses bibliothèques oblige les développeurs à reconsidérer leur rôle lorsque le logiciel évolue. Dans cette thèse, nous identifions plusieurs problématiques impliquant les bibliothèques tierces dans un contexte de maintenance logicielle. Plus particulièrement, une bibliothèque peut ne plus répondre aux besoins d’un logiciel et doit être remplacée par une nouvelle.Nous nommons cette opération une migration de bibliothèque.Nous soulevons dans ce contexte trois points qui caractérisent les difficultés rencontrées par les développeurs. Vers quelle bibliothèque migrer ? Comment appliquer la migration ? Avec l’aide de quels développeurs ? Cette thèse discute de solutions et apporte des contributions autour de ces problèmes. Nous présentons plusieurs approches et les évaluons lors de différents cas d’étude. L’analyse de l’évolution logicielle sera notre support de travail, dont la méthodologie est basée sur l’observation des changements de logiciels. Nous décrivons les limites actuelles de nos contribu-tions et ouvrons des perspectives futures pour enrichir l’état de l’art dans ce domaine / Software depend on third-party libraries to reduce development and maintenance costs. Developers have access to robust functionalities through an application programming interface designed by these libraries. However, due to the strong relationship with these libraries, developers have to reconsider their position when the software evolves. In this thesis, we identify several re-search problems involving these third-party libraries in a context of software maintenance. More specifically, a library may not satisfy the software new requirements and has to be replaced by anew one. We call this operation a library migration.We leverage three points that characterize the impediments met by developers in this situation.To which library should they migrate ? How to migrate their software ? Who can help them in this case ? This thesis suggests answers and exposes several contributions to these problems. We define three approaches that are evaluated through several case studies. To achieve this work, weuse a methodology based on software evolution analysis to observe and understand how software change. We describe numerous perspectives to overcome the current limitations of our solutions.

La thèse, réalisée dans le cadre du projet « investissements d'avenir » E-care, propose un processus méthodologique pour faciliter l'analyse et la conception de systèmes de télésurveillance médicale pour la détection précoce de signes précurseurs à toute complication. La méthodologie proposée est basée sur un système multi-agents utilisant plusieurs types d'ontologies associées à un système expert. Le système multi-agents est adapté à la télésurveillance médicale avec une architecture distribuée pour permettre l’autonomie et la réactivité au sein des sites de déploiement, en particulier les habitats. Les architectures ainsi conçues, prennent en compte l'ensemble des données du patient : son profil, ses antécédents médicaux, les traitements médicamenteux, les données physiologiques et comportementales ainsi que les données relatives à son environnement et à son hygiène de vie. Ces architectures doivent également être ouvertes pour s'adapter à de nouvelles sources de données.Cette méthodologie a été appliquée au projet E-care définissant ainsi son système d'information. Ce système d'information est composé de deux types d'ontologies représentant les connaissances pertinentes ainsi qu'un système expert pour la détection de situations à risque. Une première ontologie du problème a été construite pour gérer le système, les acteurs et leurs taches. Par la suite, trois ontologies de domaines ont été construites représentant, les maladies, les médicaments et les facteurs de risque cardio-vasculaire. Le système expert exploite des règles d'inférences construites en collaboration avec les experts médicaux et en utilisant des guides de bonnes pratiques dans le domaine de la cardiologie. Cette méthodologie a défini également l'architecture du système composé de quatre types d'agents autonomes à savoir : des capteurs pour la prise de mesures, une passerelle pour la collecte et la transmission depuis les habitats, un serveur pour le traitement et l'accès aux données, et enfin une base de données pour le stockage sécurisé des données des patients.Le système E-care a été testé et validé en utilisant des tests et des simulations inspirés de cas réels. Par la suite, une expérimentation a été faite pour la validation les différents composants du système dans milieu de télésurveillance médicale. Cette expérimentation est passée par deux phases, la première s'est déroulée au CHRU de Strasbourg, et la deuxième est en cours aux domiciles des patients. / The thesis, conducted as part of the E-care project, proposes a methodological process to facilitate the analysis and design of medical remote monitoring systems for early detection of signs of any complications. The proposed methodology is based on a multi-agent system using several types of ontologies associated with an expert system. The multi-agent system is suitable for medical monitoring with a distributed architecture to keep some autonomy and responsiveness of habitats. The process identifies the generic and specific aspects of each system. The designed architectures take into account all the patient data such as: patient profile, medical history, drug treatments, physiological and behavioral data, as well as data relating to patient's environment and his lifestyle. These architectures should be open to be adapted to new data sources.This methodology was applied to E-care project in order to define its information system. This information system is composed of two types of ontologies (problem ontology and several domain ontologies) and an expert system for the detection of risk situations. The problem ontology was built to manage the system including users and their tasks. Three domain ontologies have been built to represent, disease, drugs and cardiovascular risk factors. The expert system uses inference rules, which are defined in collaboration with medical experts using their knowledge and some medical guidelines. This methodology also defined the system architecture, which consists of four autonomous agents types namely: medical sensors to collect physiological measurements. The gateway collects data from sensors and transmits them from the patients' homes to the server. The server processes data and gives access to them. Finally the database secures storage of patient data.As part of the E-care project, an experiment was conducted to validate the various system components. This experiment has gone through two phases, the first was held at the University Hospital of Strasbourg, and the second is in the patients' homes.