Made available in DSpace on 2010-04-20T20:54:47Z (GMT). No. of bitstreams: 0 Previous issue date: 2002-02-27T00:00:00Z / Este trabalho trata do estudo da formação de preços no mercado de commodities brasileiro. O enfoque teórico fornecido pela Microstructure Theory foi utilizado juntamente com o instrumental econométrico da análise de cointegração por meio do método de Johansen. Os resultados demonstraram que o mercado brasileiro é tomador de preços. apesar de influenciar as cotações internacionais.

Made available in DSpace on 2010-04-20T20:58:06Z (GMT). No. of bitstreams: 3 DenisioAugustoLDelfino20042007.pdf.jpg: 11453 bytes, checksum: aa80c5c82fba73664fb61aa5b1a385fd (MD5) DenisioAugustoLDelfino20042007.pdf.txt: 118171 bytes, checksum: efc22da651bf354d0e820d4e5830bf61 (MD5) DenisioAugustoLDelfino20042007.pdf: 860877 bytes, checksum: 176c9b5aede2aba87c80d7594432985c (MD5) Previous issue date: 2007-04-20T00:00:00Z / The law of one price states that all identical assets, traded in different markets, must have only one price. In this dissertation, we aim to examine whether the Brazilian sovereign credit risk, traded in the international financial market, is priced similarly in the traditional bonds market as well as in the new and growing credit derivatives market. In addition to that, we make use of the Price Discovery analysis to study which of the two markets moves more rapidly in response to changes in the credit conditions in the Brazilian economy. As for the empirical analysis, we make use of time series econometrics, more specifically cointegration analysis and vector error correction. Our findings corroborate the theoretical prediction related to the law of one price, i.e., the Brazilian credit risk, either in the bonds market or in the credit derivatives market, move together in the long run. Our results also show that the majority of price discovery occurs in the credit derivatives market. / A lei do preço único afirma que o mesmo ativo negociado em diferentes mercados deve apresentar preços equivalentes. Este trabalho busca verificar se o risco de crédito soberano brasileiro negociado no mercado internacional é precificado de forma semelhante tanto nos tradicionais mercados de títulos quanto no novo e crescente mercado de derivativos de crédito. Adicionalmente, utiliza-se a análise de Price Discovery para examinar qual dos mercados se move mais rapidamente em resposta às mudanças nas condições de crédito da economia brasileira. A análise empírica é feita por meio de modelos de séries de tempo, mais especificamente análise de cointegração e vetor de correção de erros. Os resultados confirmam a predição teórica da lei do preço único de que o risco de crédito brasileiro, tanto nos mercados de títulos quanto no mercado de derivativos de crédito, movem-se juntos no longo prazo. Por fim, a maior parte do Price Discovery ocorre no mercado de derivativos de crédito.

The forms discovered by our proposal can be directly used as training data by some form classifiers. Our experimental validation used thousands of real Web forms, divided into six domains, including a representative subset of the publicly available DeepPeep form base (DEEPPEEP, 2010; DEEPPEEP REPOSITORY, 2011). Our results show that it is feasible to mitigate the demanding manual work required by two cutting-edge form classifiers (i.e., GFC and DSFC (BARBOSA; FREIRE, 2007a)), at the cost of a relatively small loss in effectiveness.

Les concepts de découverte et d’extraction de connaissances ainsi que d’inférencesont abordés sous différents angles au sein de la littérature scientifique. En effet, de nombreux domaines s’y intéressent allant de la recherche d’information, à l’implication textuelle en passant par les modèles d’enrichissement automatique des bases de connaissances. Ces concepts suscitent de plus en plus d’intérêt à la fois dans le monde académique et industriel favorisant le développement de nouvelles méthodes.Cette thèse propose une approche automatisée pour l’inférence et l’évaluation de connaissances basée sur l’analyse de relations extraites automatiquement à partir de textes. L’originalité de cette approche repose sur la définition d’un cadre tenant compte (i) de l’incertitude linguistique et de sa détection dans le langage naturel réalisée au travers d’une méthode d’apprentissage tenant compte d’une représentation vectorielle spécifique des phrases, (ii) d’une structuration des objets étudiés (e.g. syntagmes nominaux) sous la forme d’un ordre partiel tenant compte à la fois des implications syntaxiques et d’une connaissance a priori formalisée dans un modèle de connaissances de type taxonomique (iii) d’une évaluation des relations extraites et inférées grâce à des modèles de sélection exploitant une organisation hiérarchique des relations considérées. Cette organisation hiérarchique permet de distinguer différents critères en mettant en œuvre des règles de propagation de l’information permettant ainsi d’évaluer la croyance qu’on peut accorder à une relation en tenant compte de l’incertitude linguistique véhiculée. Bien qu’a portée plus large, notre approche est ici illustrée et évaluée au travers de la définition d’un système de réponse à un questionnaire, généré de manière automatique, exploitant des textes issus du Web. Nous montrons notamment le gain informationnel apporté par la connaissance a priori, l’impact des modèles de sélection établis et le rôle joué par l’incertitude linguistique au sein d’une telle chaîne de traitement. Les travaux sur la détection de l’incertitude linguistique et la mise en place de la chaîne de traitement ont été validés par plusieurs publications et communications nationales et internationales. Les travaux développés sur la détection de l’incertitude et la mise en place de la chaîne de traitement sont disponibles au téléchargement à l’adresse suivante : https ://github.com/PAJEAN/. / Knowledge discovery and inference are concepts tackled in different ways in the scientific literature. Indeed, a large number of domains are interested such as : information retrieval, textual inference or knowledge base population. Theses concepts are arousing increasing interest in both academic and industrial fields, promoting development of new methods.This manuscript proposes an automated approach to infer and evaluate knowledge from extracted relations in non-structured texts. Its originality is based on a novel framework making possible to exploit (i) the linguistic uncertainty thanks to an uncertainty detection method described in this manuscript (ii) a generated partial ordering of studied objects (e.g. noun phrases) taking into account of syntactic implications and a prior knowledge defined into taxonomies, and (iii) an evaluation step of extracted and inferred relations by selection models exploiting a specific partial ordering of relations. This partial ordering allows to compute some criteria in using information propagation rules in order to evaluate the belief associated to a relation in taking into account of the linguistic uncertainty. The proposed approach is illustrated and evaluated through the definition of a system performing question answering by analysing texts available on the Web. This case study shows the benefits of structuring processed information (e.g. using prior knowledge), the impact of selection models and the role of the linguistic uncertainty for inferring and discovering new knowledge. These contributions have been validated by several international and national publications and our pipeline can be downloaded at https ://github.com/PAJEAN/.

Made available in DSpace on 2014-06-11T19:32:49Z (GMT). No. of bitstreams: 0 Previous issue date: 2008-12-05Bitstream added on 2014-06-13T18:44:15Z : No. of bitstreams: 1 fleck_gf_dr_assis.pdf: 1496681 bytes, checksum: b239b26acf221258c987f20810cb64d4 (MD5) / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Apoiado nos pressupostos da literatura comparada, o presente estudo investiga as principais produções de romances no contexto da poética do descobrimento em terras americanas, cujas produções incluem obras que revelam desde o discurso apologético ao paródico e carnavalizado. Tais releituras se alimentam das imagens dicotômicas de Colombo, expostas tanto pela historiografia tradicional como pela nova história e pelo romance histórico. Partindo de um corpus bastante amplo, com romances produzidos em diferentes períodos, objetiva-se, primeiramente, destacar o discurso apologético presente nas produções norte-americanas desde o romantismo até a contemporaneidade para, em seguida, ao abordar a produção hispano-americana da temática – iniciada na década de 70 do século XX – mostrar que esta modalidade de novo romance histórico acabou influenciando as produções sobre o descobrimento em todo o território americano, levando determinados romancistas norteamericanos a se alinharem com os logros estéticos da literatura hispano-americana. Como conseqüência deste processo, temos, na literatura norte-americana contemporânea, um conjunto de obras que inclui a dialética da apologia e da paródia em relação à poética do descobrimento, fato que revelamos pela análise de um corpus específico. O procedimento de seleção do corpus levou em conta, para sua delimitação, os seguintes critérios: obras que pertencessem à modalidade romance histórico americano contemporâneo; que fossem oriundas de cada uma das três Américas; e que fossem representativas de cada uma das modalidades contemporâneas de romance histórico em língua espanhola e inglesa... / Supported by the principles of Compared Literature, the present study investigates the main novels’ productions on the poetry of the discovery in America, whose fictional production includes works with different discourses, from the apology to parody and carnivalization. Such re-elaborations have as one of their sources the dichotomic images of Columbus widespread by both the traditional historiography and the new history, as well as the historical novel. Based on a comprehensive corpus of historical novels produced in different periods in America, we firstly intend to prove that this literary genre in North America has produced a discourse of apology since Romanticism until contemporary times, and then, by approaching the Hispanic American fictional production of the theme – which started around the 70s of the 20th century –, to show that this kind of historical novel has eventually influenced the whole fictional production on the discovery in the whole American continent. This fact made some of the North American novelist align themselves with the aesthetic aspects achieved by the Hispanic American Literature. As a consequence of this process, the current North American Literature presents a number of works including the dialectic of apology and parody in relation to the discovery, which can be confirmed by the analyses of works from our corpus. In the selection of this corpus, we considered the following criteria: works classified as contemporary American historical novels; works produced in each of the three Americas; and works which are representative of the different... (Complete abstract click electronic access below)

Com o crescimento das empresas que fazem uso das tecnologias de bancos de dados, os administradores destes bancos de dados criam novos esquemas a cada instante, e na maioria dos casos não existe uma normalização ou procedimentos formais para que tal tarefa seja desempenhada de forma homogênea, resultando assim em bases de dados incompatíveis, o que dificulta a troca de dados entre as mesmas. Quando os Sistemas de Bancos de Dados (SBD) são projetados e implementados independentemente, é normal que existam incompatibilidades entre os dados de diferentes SBD. Como principais conflitos existentes nos esquemas de SBD, podem ser citados problemas relacionados aos nomes dos atributos, armazenamento em diferentes unidades de medida, diferentes níveis de detalhes, atributos diferentes com mesmo nome ou atributos iguais com nomes diferentes, tipos de dado diferentes, tamanho, precisão, etc. Estes problemas comprometem a qualidade da informação e geram maiores custos em relação à manutenção dos dados. Estes problemas são conseqüências de atributos especificados de forma redundante. Estes fatos têm provocado grande interesse em descobrir conhecimento em banco de dados para identificar informações semanticamente equivalentes armazenadas nos esquemas. O processo capaz de descobrir este conhecimento em banco de dados denomina-se DCDB (Descoberta de Conhecimento em Bancos de Dados). As ferramentas disponíveis para a execução das tarefas de DCDB são genéricas e derivadas de outras áreas do conhecimento, em especial, da estatística e inteligência artificial. As redes neurais artificiais (RNA) têm sido utilizadas em sistemas cujo propósito é a identificação de padrões, antes desconhecidos. Estas redes podem aprender similaridades entre os dados, diretamente de suas instâncias, sem conhecimento a priori. Uma RNA que tem sido usada com êxito para identificar equivalência semântica é o Mapa Auto-Organizável (SOM). Esta pesquisa objetiva descobrir, de modo semi-automatizado, equivalência semântica entre atributos de bases de dados, contribuindo para o gerenciamento e integração das mesmas. O resultado da pesquisa gerou uma sistemática para o processo de descoberta e uma ferramenta que a implementa. / With the increasing number of companies using database technologies, the database’s administrators create new schemes at every moment, and in most cases there are no normalization or formal procedures to do this task in a homogeneous form, it results in incompatible databases, that difficult data exchange. When the Database Systems (DBS) are projected and implemented independently, it is normal that data incompatibilities among different DBS. Problems related to the names of the attributes, storage in different measurement units, different levels of detail, different attributes with the same name or equal attributes with different names, different type of data, size, precision, etc, can be cited as main conflicts existing in the DBS schemes. These problems compromise the quality information and generate higher costs regarding the data maintenance. These problems arise as the consequence of redundant attributes’ specification. These facts have caused great interest in discovering knowledge in database to identify information semantically equivalent stored in schemes. The process capable to discover this knowledge in database is called KDD (Knowledge Discovery in Database). The available tools to do KDD tasks are generic and derived from other areas of knowledge, in special, statistics and artificial intelligence. The artificial neural networks (ANN) have been used in systems which aim is the identification of previously unknown patterns. These networks can learn similarities among the data directly from instances, without a priori knowledge. An ANN that has been used with success to identify semantic equivalence is the Self-Organizing Map (SOM). This research aims to discover, in a semi-automatic way, semantic equivalence on database attributes, contributing for the management and integration of these databases. This work resulted in a systematic for the discovery process and a tool that implements it.

Les événements indésirables (EIs) sont devenus une vraie préoccupation du monde médical, leur réduction étant recherchée pour assurer la meilleure sécurité possible pour les patients. Les EIs sont, selon la HAS, ‘‘ des situations qui s'écartent de procédures ou de résultats escomptés dans une situation habituelle et qui sont ou qui seraient potentiellement sources de dommages’’. Alors que les EIs postopératoires sont étudiés depuis de nombreuses années, ceux ayant lieu au cours des opérations ne le sont que depuis récemment, comme le montre la récente classification des EIs intraopératoires par Kaafarani et al. publié en 2014. Cependant, la classification d'EIs intraopératoires n'est que la première étape pour comprendre les comportements chirurgicaux qui les entraînent.Dans cette thèse, nous présenterons des méthodes pour détecter l'apparition de déviations dues à l'apparition d'EIs intraopératoires et pour identifier des comportements chirurgicaux à partir de modèle de processus chirurgicaux.Ce travail a nécessité de concevoir et développer une modélisation formelle de la rectopexie et des événements indésirables qui sont associés à cette procédure chirurgicale grâceà la mise en place d'ontologies. Cette modélisation formelle nous a permis de bien appréhender le principe de cette opération et de fournir un vocabulaire permettant une annotation détaillé de vidéos endoscopiques de rectopexies, afin de créer des modèles de processus chirurgicaux en jeu.Grâce à l'annotation des vidéos chirurgicales basée sur cette modélisation, nous avons développé une une méthode de détection automatique des déviations dues à l'apparition d'événements indésirables Cette méthode est basée sur un alignement temporel non-linéaire multi-dimensionnel, que nous avons développé, suivi d'un modèle semi-Markovien caché que nous avons entraîné pour déterminer s'il existe des déviations par rapport à une chirurgie de référence et si celles-ci sont dues à des événements indésirables.Cette détection de déviations dues aux événements indésirables est la première étape afin de comprendre les raisons de leurs apparitions. Nous émettons l'hypothèse que leurs apparitions peuvent être expliquées par une succession d’activités, c'est-à-dire un pattern. Pour répondre à cette hypothèse, nous avons mis en place une méthode de découverte de patterns permettant d'identifier les comportements chirurgicaux spécifiques à différents critères. Cette identification de comportements chirurgicaux est réalisée par une classification ascendante hiérarchique avec la mise en place d'une nouvelle métrique basée sur les patterns partagés entre les chirurgies. Afin de valider notre méthode, nous l'avons comparé à deux études mettant en évidence des différences de comportements chirurgicaux, comme par exemple entre différents sites chirurgicaux ou entre deux types de procédure de la même opération. Une fois la méthode validée, nous avons utilisé notre méthode afin de montrer s'il existait des comportements chirurgicaux spécifiques à des données préopératoires et à l'apparition d'événements indésirables.Pour finir, nous revenons sur les contributions les plus importantes de ces travaux à travers une discussion générale et nous proposons différentes pistes pour améliorer nos résultats / L'auteur n'a pas fourni de résumé en anglais

abstract: Relapse after tumor dormancy is one of the leading causes of cancer recurrence that ultimately leads to patient mortality. Upon relapse, cancer manifests as metastases that are linked to almost 90% cancer related deaths. Capture of the dormant and relapsed tumor phenotypes in high-throughput will allow for rapid targeted drug discovery, development and validation. Ablation of dormant cancer will not only completely remove the cancer disease, but also will prevent any future recurrence. A novel hydrogel, Amikagel, was developed by crosslinking of aminoglycoside amikacin with a polyethylene glycol crosslinker. Aminoglycosides contain abundant amount of easily conjugable groups such as amino and hydroxyl moieties that were crosslinked to generate the hydrogel. Cancer cells formed 3D spheroidal structures that underwent near complete dormancy on Amikagel high-throughput drug discovery platform. Due to their dormant status, conventional anticancer drugs such as mitoxantrone and docetaxel that target the actively dividing tumor phenotype were found to be ineffective. Hypothesis driven rational drug discovery approaches were used to identify novel pathways that could sensitize dormant cancer cells to death. Strategies were used to further accelerate the dormant cancer cell death to save time required for the therapeutic outcome. Amikagel’s properties were chemo-mechanically tunable and directly impacted the outcome of tumor dormancy or relapse. Exposure of dormant spheroids to weakly stiff and adhesive formulation of Amikagel resulted in significant relapse, mimicking the response to changes in extracellular matrix around dormant tumors. Relapsed cells showed significant differences in their metastatic potential compared to the cells that remained dormant after the induction of relapse. Further, the dissertation discusses the use of Amikagels as novel pDNA binding resins in microbead and monolithic formats for potential use in chromatographic purifications. High abundance of amino groups allowed their utilization as novel anion-exchange pDNA binding resins. This dissertation discusses Amikagel formulations for pDNA binding, metastatic cancer cell separation and novel drug discovery against tumor dormancy and relapse. / Dissertation/Thesis / Doctoral Dissertation Bioengineering 2016

abstract: One of the most common errors developers make is to provide incorrect string identifiers across the HTML5-JavaScript-CSS3 stack. The existing literature shows that a significant percentage of defects observed in real-world codebases belong to this category. Existing work focuses on semantic static analysis, while this thesis attempts to tackle the challenges that can be solved using syntactic static analysis. This thesis proposes a tool for quickly identifying defects at the time of injection due to dependencies between HTML5, JavaScript, and CSS3, specifically in syntactic errors in string identifiers. The proposed solution reduces the delta (time) between defect injection and defect discovery with the use of a dedicated just-in-time syntactic string identifier resolution tool. The solution focuses on modeling the nature of syntactic dependencies across the stack, and providing a tool that helps developers discover such dependencies. This thesis reports on an empirical study of the tool usage by developers in a realistic scenario, with the focus on defect injection and defect discovery times of defects of this nature (syntactic errors in string identifiers) with and without the use of the proposed tool. Further, the tool was validated against a set of real-world codebases to analyze the significance of these defects. / Dissertation/Thesis / Masters Thesis Computer Science 2016

Conselho Nacional de Desenvolvimento Científico e Tecnológico / The virtual learning environments (VLEs) are benefited with advances in the use of technologies in education, enabling a more dynamic and meaningful learning. In the face of increased interaction in these environments, greatly increases the amount of data stored. The process of knowledge discovery in database (KDDKnowledge Discovery in Databases) has been used successfully in several areas and in the academic area some results have been used to assist the teachers. This dissertation describes a survey conducted with the steps of KDD, which utilizes the WEKA tool (free data mining software), specifically the J48 algorithm, to apply data mining techniques on the information stored in the database, in order to detect the student performance while running the course. The research scenario was constructed with data from assessments of introduction to media in education, Integration of the specialization course in Media in education, composed of 134 (one hundred and thirty-four) students, distributed in 5 (five) different poles. In this way, with the results obtained in the research, noted that the application of rules of the algorithm, can be a valuable instrument to professor during the execution of the course, and not only a posteriori, because it allows a positive immediate intervention of even in several variables that impact on the success of the apprentice, as type of material, discussions, activities, methodologies and strategies. / Os Ambientes Virtuais de Aprendizagem (AVAs) são beneficiados com os avanços do uso de tecnologias na Educação, possibilitando uma aprendizagem mais dinâmica e significativa. Diante do aumento de interação nestes ambientes, aumenta consideravelmente o volume de dados armazenados. O processo de Descoberta de Conhecimento em Base de Dados (Knowledge Discovery in Databases - KDD) vem sendo utilizado com sucesso em diversas áreas e na área acadêmica alguns resultados têm sido utilizados para auxiliar os professores. A presente dissertação descreve uma pesquisa realizada com as etapas de KDD, que utiliza a ferramenta WEKA (software de mineração de dados livre), em específico o algoritmo J48, para aplicar técnicas de mineração de dados nas informações armazenadas no banco de dados, a fim de detectar o desempenho dos alunos durante a execução do curso. O cenário de investigação foi construído com os dados oriundos das avaliações da disciplina de Introdução à Integração de Mídias na Educação, do Curso de Especialização em Mídias na Educação, composto de 134 (cento e trinta e quatro) alunos, distribuídos em 5 (cinco) polos distintos. Dessa forma, com os resultados obtidos na pesquisa, observou se que a aplicação de regras do algoritmo, pode ser um valioso instrumento ao professor durante a execução do curso, e não apenas a posteriori, pois possibilita uma intervenção positiva imediata do mesmo, nas diversas variáveis que impactam no sucesso do aprendiz, como tipo de material, discussões, atividades, metodologias e estratégia.