Dans cette thèse, nous nous sommes intéressés au problème de la classification à « grain fin » qui est une tâche de classification particulière où les classes peuvent être visuellement distinguables seulement à partir de détails subtils et où le contexte agit souvent comme une source de bruit. Ce travail est principalement motivé par le besoin de concevoir des représentations d'images plus « fines » pour adresser de telles tâches de classification qui nécessitent un encodage d’informations discriminantes très fines et localisées. L'originalité principale de notre approche est d’intégrer dans une représentation globale de haute dimension une mesure de consistance géométrique locale entre l’image à représenter et les images d’une base de référence (que nous considérons comme un vocabulaire visuel possiblement constitué d’un grand nombre d’images). Ceci nous permet d’encoder dans une représentation vectorielle des motifs très localisés et géométriquement consistant avec l’image (contrairement aux méthodes de codage traditionnelles comme les Bag-of-Visual-Word, les vecteurs de Fisher ou les vecteurs VLAD). Plus en détails : Nous proposons dans un premier temps une approche de classification d'instances d'entités visuelles basée sur un classificateur par plus proches voisins qui agrège les similarités entre l'image requête et celles de la base d'apprentissage. Les similarités sont calculées avec prise en compte de la consistance géométrique locale entre les descripteurs locaux de la requête et ceux des images de la base d'apprentissage. Cette base pouvant être constituée de nombreux descripteurs locaux, nous proposons de passer notre méthode à l’échelle en utilisant des méthodes de recherche approximatives de plus proches voisins. Par la suite, nous avons mis au point un nouveau noyau de similarité entre des images basé sur les descripteurs locaux qu'elles partagent dans une base de référence. Nous avons nommé ce noyau Shared Nearest Neighbors Kernel (SNN Kernel), qui peut être utilisé comme n'importe quel autre noyau dans les machines à noyau. Nous avons dérivé, à partir de ce dernier, une représentation explicite globale des images à décrire. Cette représentation encode la similarité de l'image considérée avec les différentes régions visuelles des images de la base correspondant au vocabulaire visuel. Nous avons également rendu possible l'intégration de l'information de consistance géométrique dans nos représentations à l'aide de l'algorithme RANSAC amélioré que nous avons proposé dans notre contribution précédente. La classification des images se fait ensuite par un modèle linéaire appris sur ces représentations. Finalement, nous proposons, comme troisième contribution, une stratégie permettant de considérablement réduire, jusqu'à deux ordres de grandeur, la dimension de la représentation d'image sur-complète précédemment présentée tout en conservant une performance de classification compétitive aux méthodes de l’état de l’art. Nous avons validé nos approches en conduisant une série d’expérimentations sur plusieurs tâches de classification impliquant des objets rigides comme FlickrsLogos32 ou Vehicles29, mais aussi sur des tâches impliquant des concepts visuels plus finement discriminables comme la base FGVC-Aircrafts, Oxford-Flower102 ou CUB-Birds200. Nous avons aussi démontré des résultats significatifs sur des tâches de classification audio à grain fin comme la tâche d'identification d'espèce d'oiseau de LifeCLEF2015 en proposant une extension temporelle de notre représentation d'image. Finalement, nous avons montré que notre technique de réduction de dimension permet d’obtenir un vocabulaire visuel très interprétable composé des régions d'image les plus représentatives pour les concepts visuels représentés dans la base d’apprentissage. / This thesis focuses on the issue of fine-grained classification which is a particular classification task where classes may be visually distinguishable only from subtle localized details and where background often acts as a source of noise. This work is mainly motivated by the need to devise finer image representations to address such fine-grained classification tasks by encoding enough localized discriminant information such as spatial arrangement of local features.To this aim, the main research line we investigate in this work relies on spatially localized similarities between images computed thanks to efficient approximate nearest neighbor search techniques and localized parametric geometry. The main originality of our approach is to embed such spatially consistent localized similarities into a high-dimensional global image representation that preserves the spatial arrangement of the fine-grained visual patterns (contrary to traditional encoding methods such as BoW, Fisher or VLAD Vectors). In a nutshell, this is done by considering all raw patches of the training set as a large visual vocabulary and by explicitly encoding their similarity to the query image. In more details:The first contribution proposed in this work is a classification scheme based on a spatially consistent k-nn classifier that relies on pooling similarity scores between local features of the query and those of the similar retrieved images in the vocabulary set. As this set can be composed of a lot of local descriptors, we propose to scale up our approach by using approximate k-nearest neighbors search methods. Then, the main contribution of this work is a new aggregation-based explicit embedding derived from a newly introduced match kernel based on shared nearest neighbors of localized feature vectors combined with local geometric constraints. The originality of this new similarity-based representation space is that it directly integrates spatially localized geometric information in the aggregation process.Finally, as a third contribution, we proposed a strategy to drastically reduce, by up to two orders of magnitude, the high-dimensionality of the previously introduced over-complete image representation while still providing competitive image classification performance.We validated our approaches by conducting a series of experiments on several classification tasks involving rigid objects such as FlickrsLogos32 or Vehicles29 but also on tasks involving finer visual knowledge such as FGVC-Aircrafts, Oxford-Flower102 or CUB-Birds200. We also demonstrated significant results on fine-grained audio classification tasks such as the LifeCLEF 2015 bird species identification challenge by proposing a temporal extension of our image representation. Finally, we notably showed that our dimensionality reduction technique used on top of our representation resulted in highly interpretable visual vocabulary composed of the most representative image regions for different visual concepts of the training base.

Nesta tese, apresentamos nossas estratégias para desenvolver um ambiente matemático e computacional para análises em larga-escala de dados de expressão gênica obtidos pela tecnologia de microarranjos de DNA. As análises realizadas visaram principalmente à identificação de marcadores moleculares de diagnóstico e prognóstico de cânceres humanos. Apresentamos o resultado de diversas análises implementadas através do ambiente desenvolvido, as quais conduziram a implementação de uma ferramenta computacional para a anotação automática de plataformas de microarranjos de DNA e de outra ferramenta destinada ao rastreamento da análise de dados realizada em ambiente R. Programação eXtrema (eXtreme Programming, XP) foi utilizada como técnica de planejamento e gerenciamento dos projetos de análise dados de expressão gênica. Todos os conjuntos de dados foram obtidos por nossos colaboradores, utilizando-se duas diferentes plataformas de microarranjos de DNA: a primeira enriquecida em regiões não-codificantes do genoma humano, em particular regiões intrônicas, e a segunda representando regiões exônicas de genes humanos. A primeira plataforma foi utilizada para avaliação do perfil de expressão gênica em tumores de próstata e rim humanos, sendo que análises utilizando SAM (Significance Analysis of Microarrays) permitiram a proposição de um conjunto de 49 sequências como potenciais biomarcadores de prognóstico de tumores de próstata. A segunda plataforma foi utilizada para avaliação do perfil de transcritos expressos em sarcomas, carcinomas epidermóide e carcinomas epidermóides de cabeça e pescoço. As análises com sarcomas permitiram a identificação de um conjunto de 12 genes relacionados à agressividade local e metástase. As análises com carcinomas epidermóides de cabeça e pescoço permitiram a identificação de 7 genes relacionados à metástase linfonodal. / In this PhD Thesis, we present our strategies to the development of a mathematical and computational environment aiming the analysis of large-scale microarray datasets. The analyses focused mainly on the identification of molecular markers for diagnosis and prognosis of human cancers. Here we show the results of several analyses implemented using this environment, which led to the development of a computational tool for automatic annotation of DNA microarray platforms and a tool for tracking the analysis within R environment. We also applied eXtreme Programming (XP) as a tool for planning and management of gene expression analyses projects. All data sets were obtained by our collaborators using two different microarray platforms. The first is enriched in non-coding human sequences, particularly intronic sequences. The second one represents exonic regions of human genes. Using the first platform, we evaluated gene expression profiles of prostate and kidney human tumors. Applying SAM to prostate tumor data revealed 49 potential molecular markers for prognosis of this disease. Gene expression in samples of sarcomas, epidermoid carcinomas and head and neck epidermoid carcinomas was investigated using the second platform. A set of 12 genes were identified as potential biomarkers for local aggressiveness and metastasis in sarcoma. In addition, the analyses of data obtained from head and neck epidermoid carcinomas allowed the identification of 7 potential biomarkers for lymph-nodal metastases.

Submitted by Caroline Xavier (caroline.xavier@pucrs.br) on 2017-06-30T18:19:05Z No. of bitstreams: 1 DIS_GREICE_DE_CARLI_ROMAN_COMPLETO.pdf: 9835425 bytes, checksum: aa605361de91b916006af4710a54365b (MD5) / Made available in DSpace on 2017-06-30T18:19:05Z (GMT). No. of bitstreams: 1 DIS_GREICE_DE_CARLI_ROMAN_COMPLETO.pdf: 9835425 bytes, checksum: aa605361de91b916006af4710a54365b (MD5) Previous issue date: 2016-12-15 / Em fevereiro de 2001, o Manifesto ?gil foi proposto tendo como princ?pio equipes pequenas e co-localizadas. No entanto, ao longo destes 16 anos, a agilidade tamb?m foi posta em pr?tica em outros contextos, como por exemplo: equipes distribu?das e sistemas complexos, utilizando-se o termo "Desenvolvimento ?gil em Larga Escala". N?o h? uma defini??o clara e compreensiva de como a agilidade est? presente neste contexto. Assim, nosso trabalho preenche essa lacuna com o objetivo de caracterizar a agilidade no Desenvolvimento ?gil em Larga Escala. Neste trabalho, realizou-se um estudo organizado em duas fases. Na Fase 1, denominada Base Te?rica, realizamos um estudo do estado-da-arte da ?rea. Na Fase 2, denominado Estudo Emp?rico, n?s realizamos duas investiga??es: um estudo de campo em uma empresa ?gil em larga escala, para identificar o desenvolvimento durante o processo de transforma??o da empresa para esta nova abordagem e, um grupo focal, para identificar como as equipes ?geis em larga escala que v?m utilizando os m?todos ?geis o quanto se percebem em termos de aspectos de maturidade ?gil. Estes resultados contribuem para os pesquisadores e profissionais entenderem melhor como a agilidade e definida e percebida nestes grandes ambientes. O conhecimento e ?til para aqueles que querem entender como o desenvolvimento ?gil se adapta a tais ambientes e para pesquisadores com o objetivo de se aprofundar sobre o tema. / The Agile Manifesto was proposed in February 2001 having in mind small and collocated teams. However, agile has also been put in practice in other settings (e.g. large teams, distributed teams, complex systems) under the term ?Large-Scale Agile Development' (LSAD). There is no clear definition for and understanding of how agility is present in this setting. Thus, our work fills in this gap aiming to characterize agility in LSAD. We conducted a study organized in two phases. In Phase 1, named Theoretical Base, we conducted the state-of-the-art of the area. In Phase 2, named Empirical Study, we conducted two investigations: a field study in a large-scale agile company to identify how agility was developed during the transformation process of the company to this new approach, and a focus group to identify how large-scale agile teams that have been using agile for a certain while perceive themselves in terms of maturity in agile aspects. Findings contribute to researchers and professionals better understand how agility is defined and perceived in large settings. This knowledge is useful for those who want to enter the agile journey in such similar environments and for researchers aiming to further explore the topic.

Made available in DSpace on 2016-04-04T18:33:10Z (GMT). No. of bitstreams: 1 Rafael Gabriel de Oliveira Junior.pdf: 1331034 bytes, checksum: a5fd54d875d9bda182beb535237fbff0 (MD5) Previous issue date: 2013-02-27 / This project is part of the field of studies about evaluation of basic education, specifically addressing the SYSTEM OF SCHOOL PERFORMANCE EVALUATION OF THE STATE OF S?O PAULO (SARESP), an educational policy supported by governmental alignment, with the objective of systematize the state of the art of Brazilian scientific production over the SARESP, with special focus on the impacts of this evaluation on paulista school units. The question that leads this study is: beginning with governmental of an alignment existence around the Large-Scale Evaluation systems of the basic education scope, what is the state of the art of Brazilian scientific production over the SARESP, especially in relation to the impacts of this evaluation on paulista school units, based on the study of dissertations and theses produced in Brazil from 1996 to 2011? In order to find a solution for this problem, methodological procedures of bibliographic studies were chosen as the state of the art, as well as techniques of content analysis. The results show 12 types of impact that permeate school units, the majority was identified as negative, in the face of the urgent paulista teaching demands of this important tool of evaluation called SARESP. These impacts are connected to a very fragile management, especially because of the paulista government command, with the duration of this evaluation system, notably regarding the communicational forms relating to the public of the context of school. / O presente projeto insere-se no campo dos estudos sobre avalia??o da educa??o b?sica, abordando especificamente o Sistema de Avalia??o de Rendimento Escolar do Estado de S?o Paulo (SARESP), objetivando sistematizar o estado da arte da produ??o cient?fica brasileira sobre o SARESP, com foco especial nos impactos dessa avalia??o nas unidades escolares paulistas. A quest?o que norteia o presente estudo ? a seguinte: partindo da exist?ncia de um alinhamento governativo em torno dos sistemas de avalia??o em larga escala no ?mbito da educa??o b?sica, qual ? o estado da arte da produ??o cient?fica brasileira sobre o SARESP, principalmente em rela??o aos impactos dessa avalia??o nas unidades escolares paulistas, com base no estudo das disserta??es e teses produzidas no Brasil de 1996 a 2011? A fim de encontrar uma solu??o para esse problema, foram escolhidos procedimentos metodol?gicos pr?prios dos estudos bibliogr?ficos, do tipo estado da arte, assim como t?cnicas de an?lise de conte?do. Os resultados mostram 12 tipos de impacto que permeiam as unidades escolares, diagnosticados, em sua grande maioria, como negativos, frente ?s demandas prementes do ensino paulista e pressupostos dessa importante ferramenta avaliativa chamada SARESP. Tais impactos vinculam-se a uma gest?o demasiadamente fr?gil, principalmente por parte do comando do governo paulista, ao longo da vig?ncia desse sistema de avalia??o, notadamente no que diz respeito ?s formas comunicacionais concernentes aos p?blicos do contexto escolar.

Made available in DSpace on 2016-04-04T18:33:21Z (GMT). No. of bitstreams: 1 BRUNA COLOMBO.pdf: 1151495 bytes, checksum: 60ff0eeb7094c23e1a673bdf81b5d07e (MD5) Previous issue date: 2015-02-26 / Large-scale assessment of Basic Education, focused in exams of the students performance, has been used to monitor the quality of education through the development of indicators. This study aims to investigate the repercussions of the large-scale evaluations on the quality of Basic Education, considering Prova Brasil, a nationwide exam, and SARESP,restricted to the State of S?o Paulo, whose results compose IDEB and IDESP respectively according to the prerceptions of the school managers and teachers.It consists in a qualitative approach research conducted in two primary and secondary state schools in the city of Campinas/SP and uses as production techniques of empirical material the documental analysis of the schools political pedagogical projects; semi structured interviews; and focal group, in which the school managers and teachers were the participant subjects. The theoretical background of this research had as its main contributors Almerindo Afonso, Jos? Dias Sobrinho and Luiz Carlos de Freitas, all of which excel in the field of educational assessment. It was verified that the public policies of evaluation increased in Brazil in the 1990s decade within the background of political reforms and with the consolidation of the Evaluator State. It is in this scenario that Prova Brasil and SARESP are situated, centered in students performance, whose predominant evaluation paradigm has been the prioritization of results. The perception analysis of the surveyed subjects revealed that despite the recognition regarding the gains and difficulties linked to these evaluations, the polarization of pedagogical work prevails, which manifest itself on the students preparation for the assessments in favor of rising rates. This practice demonstrates a tendency for using large-scale evaluations with regulation and control functions through quality indicators in the light of IDEB and IDESP. The participants also highlighted challenges related to intra and extraschool conditions involving teaching and learning processes. This study points to the necessity of redefining public assessment policies, with emancipatory nature, so that it transcends the mere achievement of goals translated into indicators, putting into perspective the construction of a education with social quality for everyone. / As avalia??es em larga escala da Educa??o B?sica, centradas em exames de desempenho dos alunos, t?m sido utilizadas para monitoramento da qualidade da educa??o por meio da gera??o de indicadores. O presente trabalho busca investigar quais s?o as percep??es de professores e gestores de escolas p?blicas estaduais do munic?pio de Campinas (SP) sobre as repercuss?es da Prova Brasil e do SARESP na constru??o da qualidade da educa??o, considerando a Prova Brasil, de abrang?ncia nacional, e o SARESP, restrito ao estado de S?o Paulo, cujos resultados integram o IDEB e o IDESP, respectivamente. Trata-se de uma pesquisa de abordagem qualitativa realizada em duas escolas p?blicas estaduais de ensino fundamental e ensino m?dio do munic?pio de Campinas/SP, tendo como t?cnicas de produ??o de material emp?rico a an?lise documental dos projetos pol?ticos pedag?gicos das escolas pesquisadas; a entrevista semiestruturada; e o grupo focal, sendo as gestoras escolares e os professores os sujeitos participantes. O aporte te?rico que balizou esta pesquisa teve como principais contribuintes, que se destacam no campo da avalia??o educacional, Almerindo Afonso, Jos? Dias Sobrinho e Luiz Carlos de Freitas. Constatou-se que as pol?ticas p?blicas de avalia??o, no Brasil, intensificaram-se na d?cada de 1990, no contexto das reformas pol?ticas, com a consolida??o do Estado Avaliador. ? nesse cen?rio que se inscrevem a Prova Brasil e o SARESP, centradas no desempenho dos alunos, cujo paradigma avaliativo predominante tem sido o que prioriza os resultados. A an?lise da percep??o dos sujeitos pesquisados revelou que, apesar do reconhecimento sobre os ganhos e dificuldades atrelados a essas avalia??es, prevalece a polariza??o do trabalho pedag?gico que se expressa pela prepara??o dos alunos para as avalia??es, em prol da eleva??o dos ?ndices. Tal pr?tica evidencia a tend?ncia de utiliza??o das avalia??es em larga escala com fun??o de regula??o e controle, por meio de indicadores de qualidade, como o do IDEB e do IDESP. Os participantes acentuaram, ainda, desafios que se referem a condicionantes intra e extraescolares que implicam os processos de ensino e aprendizagem. A pesquisa aponta para a necessidade de se redefinir pol?ticas p?blicas de avalia??o, com cunho emancipat?rio, que transcenda o mero alcance de metas traduzidas por meio de indicadores, perspectivando a constru??o de uma educa??o de qualidade social para todos.

Desenvolver um sistema de distribuição de energia confiável é certamente um desafio inerente aos profissionais do setor elétrico. Porém, os sistemas estão sujeitos a falhas e, sendo assim, o rápido restabelecimento traz a satisfação do cliente e reduz as compensações pagas pelas concessionárias de energia. Neste contexto a presente dissertação visa avaliar a metodologia para restabelecimento de energia denominada AEMT-H que se baseia em algoritmos evolutivos multi-objetivo se na estrutura de dados chamada Representação Nó- Profundidade (RNP). Esta avaliação ocorrerá através da aplicação do AEMT-H para obtenção de planos de restabelecimento de energia após a ocorrência de faltas simples no sistema COPEL da cidade de Londrina. Os resultados gerados serão avaliados estatisticamente e ainda subjetivamente pelos profissionais do COD da concessionária. Os algoritmos evolutivos têm apresentado resultados animadores para os problemas de restabelecimento de energia. Em especial, os resultados obtidos, quando da representação computacional de sistemas de distribuição de grande porte (com milhares de barras e chaves) através da RNP, possuem como característica o tempo de resposta da ordem de segundos, instigando a evolução das pesquisas para utilização desta metodologia inclusive em aplicativos de tempo real. O sistema de Londrina possui 30.156 barras, 2.660 chaves \"NF\", 250 chaves \"NA\" e atende um universo de mais de 231.000 consumidores ligados diretamente ao sistema de 13,8 kV ou, após os transformadores de distribuição, nas tensões de 220V e 127V. Neste sistema estão presentes 6 subestações 138 kV/13,8 kV e 64 circuitos alimentadores, totalizando uma capacidade de transformação de energia de 541,7 MVA. / The development of a reliable distribution system is certainly a challenge to electrical industry professionals. However, these systems are subject to failures and thus the fast restoration brings customer satisfaction and reduces the compensation paid by the electricity utilities. In this context, this dissertation aims to evaluate the methodology for service restoration called \"AEMT-H\", which is based on multi-objective evolutionary algorithms and in the data structure called Node Depth Encoding (NDE). This evaluation will occur by applying the \"AEMT-H\" to obtain service restoration plans considering the occurrence of simple faults in the COPEL system of the city of Londrina. The obtained results will bee valuated statistically and subjectively by professionals of the Distribution Operation Center. It is important to highlight that Evolutionary algorithms have shown promising results to treat the service restoration problem in distribution systems. In particular, the results obtained when using NDE tocomputationally represent the electrical topology of large distribution systems (with thousand of buses and switchers) are very interesting in terms of time processing (in the order of seconds). The system of Londrina has 30,156 buses, 2,660 switchers normally closed, 250 switchers normally opened and supplies a universe of more than 231,000 consumers connected directly to the system with voltage of 13.8 kV or, after distribution transformers, with voltages of 220V and 127V. This system contains 6 substations 138 kV/13,8 kV and 64 feeders, and a installed power of 541.7 MVA.

Neste trabalho, propomos um método de pontos interiores para minimização com restrições lineares de grande porte. Este método explora a linearidade das restrições, partindo de um ponto viável e preservando a viabilidade dos iterandos. Apresentamos os principais resultados de convergência global, além de uma descrição rica em detalhes de uma implementação prática de todos os passos do método. Para atestar a implementação do método, exibimos uma ampla experimentação numérica, e uma análise comparativa com métodos bem difundidos na comunidade de otimização contínua. / In this work, we propose an interior-point method for large-scale linearly constrained optimization. This method explores the linearity of the constraints, starting from a feasible point and preserving the feasibility of the iterates. We present the main global convergence results, together with a rich description of the implementation details of all the steps of the method. To validate the implementation of the method, we present a wide set of numerical experiments and a comparative analysis with well known softwares of the continuous optimization community.

Loading of data from slow persistent memory (disk storage) to main memory represents a bottleneck for current interactive visual data exploration applications, especially when applied to huge volumnes of data. Semantic caching of queries at the client-side is a recently emerging technology that can significantly improve the performance of such systems, though it may not in all cases fully achieve the near real-time responsiveness required by such interactive applications. We hence propose to augment the semantic caching techniques by applying prefetching. That is, the system predicts the user's next requested data and loads the data into the cache as a background process before the next user request is made. Our experimental studies confirm that prefetching indeed achieves performance improvements for interactive visual data exploration. However, a given prefetching technique is not always able to correctly predict changes in a user's navigation pattern. Especially, as different users may have different navigation patterns, implying that the same strategy might fail for a new user. In this research, we tackle this shortcoming by utilizing the adaptation concept of strategy selection to allow the choice of prefetching strategy to change over time both across as well as within one user session. While other adaptive prefetching research has focused on refining a single strategy, we instead have developed a framework that facilitates strategy selection. For this, we explored various metrics to measure performance of prefetching strategies in action and thus guide the adaptive selection process. This work is the first to study caching and prefetching in the context of visual data exploration. In particular, we have implemented and evaluated our proposed approach within XmdvTool, a free-ware visualization system for visually exploring hierarchical multivariate data. We have tested our technique on real user traces gathered by the logging tool of our system as well as on synthetic user traces. Our results confirm that our adaptive approach improves system performance by selecting a good combination of prefetching strategies that adapts to the user's changing navigation patterns.

Submitted by Maicon Juliano Schmidt (maicons) on 2015-03-30T19:00:05Z No. of bitstreams: 1 Ivan dos Santos Oliveira.pdf: 1254978 bytes, checksum: f10b901aced69e025f900284024bf102 (MD5) / Made available in DSpace on 2015-03-30T19:00:05Z (GMT). No. of bitstreams: 1 Ivan dos Santos Oliveira.pdf: 1254978 bytes, checksum: f10b901aced69e025f900284024bf102 (MD5) Previous issue date: 2013-10-24 / IFPI - Instituto Federal de Educação Ciência e Tecnologia do Piauí / Tendo em vista a importância que política pública de avaliação do Novo ENEM tem tomado no cenário educacional nacional, essa pesquisa procurou como objetivos principais, entender através da perspectiva dos estudantes que participaram do Novo ENEM, os sentidos e significados que os mesmos conferem à política e à sua participação nela, bem como entender a influência dessa política e de outras avaliações em larga escala em suas experiências de escolarização. Para tanto, foram utilizados como marcos conceituais de referência o ciclo de políticas (policycycle approach) de Stephen J. Ball, que serviu como base para o entendimento dos mecanismos de funcionamento dos processos que envolvem as políticas de maneira geral, considerando como importante a ação dos sujeitos como um aspecto crucial para a compreensão delas e também a Teoria da Análise de Discurso fundamentada por Michel Pêcheux, como teoria basilar da análise a fim de entender os sentidos e significados construídos pelos colaboradores da pesquisa por meio da linguagem. Através da análise dos dados produzidos nas entrevistas foi possível perceber que a falta de comunicação e de diálogo foi uma constante no ambiente escolar dos colaboradores da pesquisa, sendo este, em seus casos, o ambiente configurado para a aplicação de avaliações externas. Evidenciou-se a falta de percepção quanto à influência das avaliações externas em larga escala pelos colaboradores da pesquisa, em seus ambientes escolares. Com relação às influências e repercussões do Novo ENEM no ambiente escolar, foi verificada a existência de duas realidades distintas, uma para as escolas públicas e outra para as escolas privadas. Quanto aos sentidos e funções atribuídos pelos estudantes à política do Novo ENEM, constatou-se que o discurso oficial exerce forte influência e se faz presente nas percepções dos colaboradores quando estes evocam como sentidos para a política: a autoavaliação, a democratização na participação do processo, a indução de mudanças no ensino médio e o ingresso no ensino superior. Finalmente, foi possível entender que, mesmo para os estudantes que avaliam positivamente sua participação nesse processo, muitas vezes essa experiência pode se configurar como cansativa, desagradável e por vezes traumática, por conta de fatores que atuam no pré-prova, durante a aplicação da prova e no pós-prova. / Bearing in mind the importance that evaluation public policy of the new ENEM (High School National Exam) has achieved, this research aims – as main objectives, through the point of view of the students who had taken the exam, to comprehend what their senses and meanings about the new ENEM are. As well as understanding this policy influence and other large scale evaluations in their school experiences. For this purpose, the policy cycle approach of Stephen J. Ball was taken as reference theoretical patterns, which were bases to realize the execution mechanisms of the processes involving policies in general meanings, considering important the action exercised by the subjects as a crucial aspect for its understanding along with the Discourse Analysis Theory justified by Michel Pêcheux, as the base theory of the analysis in order to be aware of the senses and meanings built by the research collaborators through the language. By analyzing the data produced during the interviews it was possible to realize that lack of communication and dialog was frequent within the collaborators school environment, making of these the atmosphere set up to be applied foreign evaluations. It was evinced the lack of perception in terms of the influence of the foreign evaluations in a huge level by the research collaborators inside their school environment. Regarding the influence and repercussion of the new ENEM within the school surroundings, it was possible to verify the existence of two distinct realities, one for each: public and private school. Concerning the senses and functions ascribed by the students to the policy of the new ENEM, it was acknowledged that the official discourse exerts strong influence and is present in the collaborators perception when they evoke a political sense: the self-evaluation, the process participation democratization, the prompting for high school and superior education admission changes. Finally, it was possible to comprehend that, even for students that estimate their participation in this process as positive, many times these experiences may configure as wearing, unpleasant and many times traumatic, because of factors that take action before the exam, throughout the application and subsequent to the test.

Cette thèse veut enquêter sur la diffusion du discours sur la durabilité urbaine en France et sur son rôle dans le projet de ville et de territoire : l'émergence du développement durable a été abordée comme une problématique ouverte, plutôt qu'une série de solutions fermées. Notre posture critique nous permet d'enquêter la notion de durabilité d'un point de vue qui est profondément lié à la réflexion et la praxis du projet de territoire ; nous nous sommes donc centrés sur les cultures disciplinaires, en nous penchant en particulier sur les acteurs de la conception. La question de la consistance et de l'utilité de la notion de figure nous a permis d'effectuer une mise en perspective des questions actuelles, notamment par rapport au débat italien sur les outils critiques du projet. La recherche que nous avons mené semble nous indiquer que la question de la « métropole durable » est en train de modifier les équilibres et le sens des différentes figures à l'oeuvre dans le territoire de Strasbourg, dans ses représentations aussi bien que dans les récits des acteurs. / This dissertation aims at exploring on the diffusion of sustainability in France and its role in urban and large-scale design. To this aim, the arising of sustainable urban development is regarded as a set of problems, rather than a set of solutions. Our critical position allows us to challenge the notion of sustainability from a viewpoint that is profoundly linked to reflection and practice of the large-scale project; we therefore have focused on disciplinary cultures, particularly those concerning design and planning stakeholders. The notion of figure, the issue of its consistency and relevance has allowed us to put into perspective contemporary topics, especially referring to the Italian debate on tools of design thinking. Concerning the Strasbourg case-study, our research suggests that the “sustainable metropolis” is changing meanings and relationships between different figures operating in its territory and its representations, as well as its narratives. / Questa tesi vuole, in prima istanza, raccontare la diffusione della sostenibilità in Francia, dal punto di vista della sua “declinazione urbana”, e indagare il suo ruolo all'interno del progetto urbanistico e territoriale. Viene quindi presentata l'emergenza dello sviluppo sostenibile, non tanto come una soluzione, quanto come una problematica: vengono messe in luce le sue contraddizioni, le questioni che esso pone, le tensioni tra i suoi elementi costitutivi. La scelta di una postura critica permette di indagare la nozione di sostenibilità da un punto di vista profondamente legato alla riflessione e alla pratica del progetto di territorio; ci siamo quindi concentrati sulle culture disciplinari, in particolare sugli attori del progetto. La questione della consistenza e dell'utilità della nozione di figura ha costituito per noi il mezzo per riallacciarci al dibattito italiano sugli strumenti critici del progetto, interrogandone la pertinenza in una prospettiva contemporanea. La ricerca effettuata sembra indicare che la questione della métropole durable a Strasburgo stia modificando gli equilibri e i significati delle diverse figure operanti sul territorio, al livello delle sue rappresentazioni e del racconto degli attori