Nous proposons un modèle de vérification grammaticale automatique gauche-droite issu de l'analyse d'un corpus d'erreurs tapuscrites. Les travaux menés en psychologie cognitive ont montré que le processus de révision procède au travers de la confrontation d'une attente à un résultat. Ainsi, la détection d'une erreur grammaticale reposerait, chez l'humain, sur une attente du réviseur non comblée. Ce principe est à la base du modèle que nous avons élaboré. Pour faciliter la gestion des attentes du point de vue traitement numérique, nous convions deux concepts courants en TAL : le principe d'unification et la segmentation en chunks. Le premier est particulièrement adapté à la vérification des accords et le second constitue une unité de calcul intermédiaire permettant de définir des bornes simplifiant la recherche d'incohérences grammaticales. Enfin, l'originalité de ce modèle réside dans une analyse gauche-droite construite au fur et à mesure de la lecture/écriture. / This thesis presents a model for automated left-right grammar checking based on analysis of a corpus of typescript errors. Studies in cognitive psychology have shown that the revision process works by confronting expectations with results. For humans, detecting a grammatical error therefore relies on an unfulfilled expectation on the part of the revisor. The model presented here is based on this principle. In order to deal with expectations from the point of view of computational processing, two common concepts in NLP are called upon: the unification principle and chunk segmentation. The former is particularly adapted to checking agreements, while the latter provides an intermediate computational unit to delimit, and therefore simplify, detection of grammatical inconsistencies. Finally, the model?s originality lies in the left-right analysis it provides, which is constructed as the text is produced/read.

Orientador: Antonio Maria Garcia Tommaselli / Banca: Aluir Porfírio Dal Poz / Banca: Edson Aparecido Mitishita / Resumo: As diferenças radiométricas entre áreas homólogas de imagens aéreas digitais são causadas por vários fatores e afetam os processos fotogramétricos, como a correspondência e a mosaicagem. Para que as imagens possam ser melhoradas, alguns procedimentos como correção geométrica e radiométrica são aplicados, sendo a correção radiométrica o tema de estudo deste trabalho. O processo de correção radiométrica em imagens aéreas digitais tem como principal objetivo tornar semelhantes os histogramas de imagens que possuam sobreposição, facilitando os processos posteriores de correspondência de pontos homólogos e a formação de mosaicos. O método proposto neste trabalho realiza em duas etapas a correção radiométrica para um bloco de imagens aéreas digitais. A primeira etapa é um processamento global, baseado nos dados dos histogramas. A segunda etapa é uma correção local, que pode ser realizada através de dois métodos estudados, sendo que o primeiro trata a interpolação dos valores de correção à partir de regiões homólogas, possibilitando a correção radiométrica da região de sobreposição das imagens; e o segundo método utiliza o ajuste de uma superfície parabolóide às diferenças radiométricas entre imagens, fornecendo, posteriormente, valores de correção para toda a imagem. Para realizar os processos descritos acima, foi implementado um programa em linguagem C, o qual realiza o ajuste local a partir de pontos correspondentes. Alguns experimentos foram realizados, utilizando três blocos de imagens aéreas obtidas com câmeras digitais Hasselblad de 22 e 39 megapixels, respectivamente, os quais apresentaram resultados satisfatórios. / Abstract: Radiometric differences between homologous areas in digital aerial imagens are caused by several factors and affect the photogrammetric processes like correspondence and mosaicking. In order to enhance those images, some procedures like geometric and radiometric corrections can be applied, the last being the aim of this work. The process of radiometric adjustment in digital aerial images aim at to make similar the histograms of overlapping images, favoring the processes of correspondence and mosaicking formation. The method proposed in this work is performed in two stages: firstly a global processing based on the histograms data and secondly a local strategy, for which two methods were studied. The first is an interpolation of correction values corresponding to the radiometric differences in the overlapping images. The second method for the local adjustment uses surface adjustment in order to correct the differences in the corresponding regions. In order to assess the proposed methodology, a program in C language was implemented to perform the local adjustment based on some corresponding points. Some experiments were accomplished, using three blocks of aerial images taken with 22 and 39 megapixels digital cameras, and good results were achieved. / Mestre

The main purpose of this study is to examine the causal relations between energy use, CO2 emissions and economic growth for Sweden. Vector Error Correction model with annual data from 1970 to 2016 has been used in order to determine potential causality between the variables. The empirical findings indicate that in the long-run, causality relationship between energy consumption, CO2 emissions and economic growth cannot be rejected and it is bidirectional. This means that energy is a determining factor for economic growth in Sweden and that applying policies in order to reduce the CO2 emissions has slowed down economic growth in Sweden. This finding is consistent with the Feedback Hypothesis. But in the short-run no causality was found between energy and economic growth. According to Granger causality test results, bidirectional causality between CO2 emissions and energy consumption cannot be rejected in the short-run. Variables’ trends show that in the period under study, energy consumption and economic growth have moved in the same direction; meaning that higher energy consumption has led to higher economic growth. At the same time, lower CO2 emissions have been accompanied by higher economic growth. There is also short-run causality running from capital to economic growth according to VECM results. It can be suggested to the policy makers that in order to maintain economic growth and reduce environmental degradation, energy consumption should be shifted gradually from nonrenewable sources to renewable ones so to avoid decrease in economic growth and ensure lower levels of CO2 emissions in the long-run.

Made available in DSpace on 2014-06-11T19:22:25Z (GMT). No. of bitstreams: 0 Previous issue date: 2009-03-26Bitstream added on 2014-06-13T20:28:44Z : No. of bitstreams: 1 langhi_pjp_me_prud.pdf: 4558873 bytes, checksum: 6633a0b385f223c9200a10403ba4fa9b (MD5) / Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / As diferenças radiométricas entre áreas homólogas de imagens aéreas digitais são causadas por vários fatores e afetam os processos fotogramétricos, como a correspondência e a mosaicagem. Para que as imagens possam ser melhoradas, alguns procedimentos como correção geométrica e radiométrica são aplicados, sendo a correção radiométrica o tema de estudo deste trabalho. O processo de correção radiométrica em imagens aéreas digitais tem como principal objetivo tornar semelhantes os histogramas de imagens que possuam sobreposição, facilitando os processos posteriores de correspondência de pontos homólogos e a formação de mosaicos. O método proposto neste trabalho realiza em duas etapas a correção radiométrica para um bloco de imagens aéreas digitais. A primeira etapa é um processamento global, baseado nos dados dos histogramas. A segunda etapa é uma correção local, que pode ser realizada através de dois métodos estudados, sendo que o primeiro trata a interpolação dos valores de correção à partir de regiões homólogas, possibilitando a correção radiométrica da região de sobreposição das imagens; e o segundo método utiliza o ajuste de uma superfície parabolóide às diferenças radiométricas entre imagens, fornecendo, posteriormente, valores de correção para toda a imagem. Para realizar os processos descritos acima, foi implementado um programa em linguagem C, o qual realiza o ajuste local a partir de pontos correspondentes. Alguns experimentos foram realizados, utilizando três blocos de imagens aéreas obtidas com câmeras digitais Hasselblad de 22 e 39 megapixels, respectivamente, os quais apresentaram resultados satisfatórios. / Radiometric differences between homologous areas in digital aerial imagens are caused by several factors and affect the photogrammetric processes like correspondence and mosaicking. In order to enhance those images, some procedures like geometric and radiometric corrections can be applied, the last being the aim of this work. The process of radiometric adjustment in digital aerial images aim at to make similar the histograms of overlapping images, favoring the processes of correspondence and mosaicking formation. The method proposed in this work is performed in two stages: firstly a global processing based on the histograms data and secondly a local strategy, for which two methods were studied. The first is an interpolation of correction values corresponding to the radiometric differences in the overlapping images. The second method for the local adjustment uses surface adjustment in order to correct the differences in the corresponding regions. In order to assess the proposed methodology, a program in C language was implemented to perform the local adjustment based on some corresponding points. Some experiments were accomplished, using three blocks of aerial images taken with 22 and 39 megapixels digital cameras, and good results were achieved.

Made available in DSpace on 2014-06-11T19:24:49Z (GMT). No. of bitstreams: 0 Previous issue date: 2007-09-10Bitstream added on 2014-06-13T19:11:34Z : No. of bitstreams: 1 buranello_lva_me_bauru.pdf: 1083474 bytes, checksum: 25ccdb14b48213f528ada08bfe779485 (MD5) / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Secretaria da Educação de SP / O objetivo do presente estudo foi investigar as contribuições que as classes de Correção de Fluxo proporcionaram aos alunos em termos de solução de problemas matemáticos, sob a visão de alunos, professores e ATP. Para tanto foram investigados: alunos egressos das classes de Correção de Fluxo, alunos que freqüentaram apenas classes regulares, professoras que lecionaram no projeto em questão e que hoje trabalham com os alunos egressos e o assistente técnico pedagógico que realizou as capacitações durante o desenvolvimento do projeto. Trata-se de uma amostra por conveniência. Foi possível constatar através de uma análise fenomenológica que o Projeto Correção de Fluxo não conseguiu atingir seus objetivos, ou seja, não promoveu a inclusão dos alunos que dele fizeram parte, pois os mesmos não resolvem problemas básicos envolvendo os números inteiros. Foi possível verificar que a exclusão destes alunos apenas foi adiada na instituição escolar investigada. Tais resultados apontam para a necessidade de melhorias em termos de políticas públicas educacionais, assim como de suas medidas operacionais. / The objetive of the present study was investigate the contributions that the classes of Correction of Flow provided to the students in terns of solution of mathematical problems, under the student`s vision, teacher and TPA. For so much whe investigated: regressor students of the classes of “Flow Correction”, students that just frequented regular classes, teachers that taught in the project in subject and that today work with the egressor students and the pedagogia technical assistant that realized the trainings during the development of the project. It is a sample for convenience. It was possibible to verify through a fenomenologic analysis that the project “Flow correction” didn’t get to reach its objectives, that is to say, it didn’t promote the student’s inclusion that took past of its, because the same ones don’t solve basic problems, involving the whole numbers. It was possible to verify that these student’s exclusion was just postponed in the investigated school institution. Such results point for the need of improvements in tenns of educational public politics, as wele as for their operational measures.

Synchrocyclotrons and laser based particle accelerators, developed with the goal to enable more compact particle therapy facilities, may bring highly pulsed radiation field to external beam radiation therapy. In addition, such highly pulsed fields may be desirable due to their potential clinical benefits regarding better healthy tissue sparing or improved gating for moving tumors. However, they pose new challenges for dosimetry, the corner stone of any application of ionizing radiation. These challenges affect both clinical and radiation protection dosimetry. Air-filled ionization chambers, which dominate clinical dosimetry, face the problem of increased signal loss due to volume recombination when a highly pulsed field liberates a large amount of charge in a short time in the chamber. While well established descriptions exist for this volume recombination for the moderately pulsed fields in current use (Boag's formulas), the assumptions on which those descriptions are based will most likely not hold in the prospective, highly pulsed fields of future accelerators. Furthermore, ambient dose rate meters used in radiation protection dosimetry as survey meters or fixed installations are generally only tested for continuous fields, casting doubt on their suitability to measure pulsed fields. This thesis investigated both these aspects of dosimetry - clinical as well as radiation protection - to enable the medical application of highly pulsed radiation fields. For a comprehensive understanding, experimental investigations were coupled with theoretical considerations and developments. Pulsed fields, varying in both dose-per-pulse and pulse duration over a wide range, were generated with the ELBE research accelerator, providing a 20 MeV pulsed electron beam. Ionization chambers for clinical dosimetry were investigated using this electron beam directly, with an aluminium Faraday cup providing the reference measurement. Whereas the dose rate meters were irradiated in the photon field generated from stopping the electron beam in the Faraday cup. In those measurements, the reference was calculated from the ionization chamber, then serving a an electron beam monitor, cross-calibrated to the photon field with thermoluminescent dosimeters. Three dose rate meters based on different operating principles were investigated, covering a large portion of the operating principles used in radiation protection: the ionization chamber based RamION, the proportional counter LB 1236-H10 and the scintillation detector AD-b. Regarding clinical dosimetry, measurements of two prominent ionization chamber geometries, plane-parallel (Advanced Markus chamber) and thimble type (PinPoint chamber), were performed. In addition to common air-filled chambers, chambers filled with pure nitrogen and two non-polar liquids, tetramethylsilane and isooctane, were investigated. In conjunction with the experiments, a numerical solution of the charge liberation, transport, and recombination processes in the ionization chamber was developed to calculate the volume recombination independent of the assumptions necessary to derive Boag's formulas. Most importantly, the influence of the liberated charges in the ionization chamber on the electric field, which is neglected in Boag's formulas, is included in the developed calculation. Out of the three investigated dose rate meters only the RamION could be identified as an instrument truly capable of measuring a pulsed field. The AD-b performed below expectations (principally, a scintillator is not limited in detecting pulsed radiation), which was attributed to the signal processing, emphasizing the problem of a typical black-box signal processing in commercial instruments. The LB 1236-H10, on the other hand, performed as expected of a counting detector. While this supports the recent effort to formalize these expectations and standardize testing for counting dosimeters in DIN IEC/TS 62743, it also highlights the insufficiency of counting detectors for highly pulsed fields in general and shows the need for additional normative work to establish requirements for dose rate meters not based on a counting signal (such as the RamION), for which no framework currently exists. With these results recognized by the German radiation protection commission (SSK) the first steps towards such a framework are taken. The investigation of the ionization chambers used in radiation therapy showed severe discrepancies between Boag's formulas and the experimentally observed volume recombination. Boag's formulas describe volume recombination truly correctly only in the two liquid-filled chambers. All the gas-filled chambers required the use of effective parameters, resulting in values for those parameters with little to no relation to their original meaning. Even this approach, however, failed in the case of the Advanced Markus chamber for collection voltages ≥ 300 V and beyond a dose-per-pulse of about 100 mGy. The developed numerical model enabled a much better calculation of volume recombination and allowed the identification of the root of the differences to Boag's formulas as the influence of the liberated charges on the electric field. Increased positive space charge due to increased dose-per-pulse slows the collection and reduces the fraction of fast, free electrons, which are unaffected by volume recombination. The resultant increase in the fraction of charge undergoing volume recombination, in addition to the increase in the total amount of charge, results in an increase in volume recombination with dose-per-pulse that is impossible to describe with Boag's formulas. It is particularly relevant in the case of high electric fields and small electrode distances, where the free electron fraction is large. In addition, the numerical calculation allows for arbitrary pulse durations, while Boag's formulas apply only to very short pulses. In general, the numerical calculation worked well for plane-parallel chambers, including those filled with the very diverse media of liquids, nitrogen and air. Despite its increased complexity, the thimble geometry could be implemented as well, although, in the case of the PinPoint chamber, some discrepancies to the experimental data remained, probably due to the required geometrical approximations. A possible future development of the numerical calculation would be an improved description of the voltage dependence of the volume recombination. At the moment it requires characterizing a chamber at each desired collection voltage, which could be eliminated by an improved modeling of the volume recombination's dependence on collection voltage. Nevertheless, the developed numerical calculation presents a marked improvement over Boag's formulas to describe the dose-per-pulse dependence and pulse duration dependence of volume recombination in ionization chambers, in principle enabling the application of ionization chambers in the absolute dosimetry of highly pulsed fields.

No description available.

Orientador: João Francisco Galera Monico / Coorientador: Milton Hirokazu Shimabukuro / Coorientador: : Marcio de Oliveira Aquino / Banca: Celso Braga de Mendonça / Banca: Hélio Koiti Kuga / Banca: Silvio Jacks dos Anjos Garnés / Resumo: Dentre os diversos métodos possíveis de posicionamento GNSS, encontra-se atualmente em evidência aqueles que proporcionam posicionamento em tempo real (ou próximo do tempo real) com acurácia ao nível centimétrico ou decimétrico. Nesse caso, destaca-se o PPP em tempo real, o qual requer a disponibilização de órbitas precisas em tempo real, bem como as correções ou erros dos relógios dos satélites. Atualmente, é possível utilizar as efemérides preditas denominadas IGU, as quais são disponibilizadas pelo IGS, porém, as correções dos relógios dos satélites contidas nessas efemérides não apresentam acurácia suficiente (3 ns @ 0,9 m) para obtenção de resultados com qualidade centimétrica no PPP. Logo, é necessário pesquisar e desenvolver metodologias adequadas para estimar as correções dos relógios dos satélites com melhor qualidade para aplicação no PPP em tempo real. A estimativa das correções dos relógios dos satélites pode ser realizada com base numa rede de estações GNSS de referência. Nesta tese, a metodologia adotada para estimar estas correções consiste em realizar o ajustamento dos dados no modo PPP para uma rede de estações GNSS. Dessa forma, todos os efeitos sistemáticos envolvidos com os sinais dos satélites GNSS devem ser modelados adequadamente para cada estação da rede, realizando-se assim a estratégia denominada de PPP em rede. Uma vez estimadas as correções dos relógios dos satélites em tempo real, estas devem ser enviadas ao usuário, o qual as utilizará para aplicação... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: Among several possible methods of GNSS positioning, it is currently in evidence those that provide real time positioning (or near real time) with accuracy near to decimeter or centimeter level. In this case, we highlight the real time PPP method, which requires the availability of real time precise orbits and corrections or errors of the satellites clocks. Currently, it is possible to use the predicted IGU ephemerides which are made available by the IGS centers. However, the satellites clocks corrections available in the IGU do not present enough accuracy (3 ns @ 0.9 m) to accomplish real time PPP with the level of centimeter accuracy. Therefore, it is necessary to research and develop appropriate methodologies for estimating the satellite clock corrections in real time with better quality for real time PPP aplication. The estimation of satellite clock corrections can be performed based on a GNSS network of reference. In this PhD thesis, the methodology used to estimate these corrections is based on the adjustment of data in the PPP mode for stations of a GNSS network. Thus, all systematic effects involved with the GNSS satellite signals must... (Complete abstract click electronic access below) / Doutor

Orientador: Leonardo Theodoro Büll / Banca: Francisco Maximino Fernandes / Banca: Renato de Melo Prado / Resumo: A eficiência dos corretivos de acidez do solo, e, sua qualidade depende da taxa de reatividade (RE), e do poder de neutralização (PN), sendo esta eficiência indicada pelo poder relativo de neutralização total (PRNT). Para os silicatos existem poucos estudos que estabeleça taxas de reatividade para suas frações granulométricas e, portanto, utilizam-se as mesmas taxas de reatividade do calcário. Foram desenvolvidos três experimentos no período de julho/2008 a outubro/2009 em casa de vegetação na Faculdade de Ciências Agronômicas/UNESP - Botucatu-SP, com o objetivo de comparar valores de reatividade das frações granulométricas para materiais silicatados em três solos de diferentes classes texturais cultivados com plantas de alfafa. Utilizou-se seis materiais corretivos: escória, silicato de cálcio, silicato de cálcio e magnésio, wollastonita, calcário dolomítico e calcário calcítico os quais no experimento 1 e 2 foram separados em quatro frações granulométricas, peneiras: nº 10 (Ø = 2 mm), nº 20 (Ø = 0,84 mm), nº 50 (Ø = 0,30 mm) e fundo (Ø<0,30 mm), exceto a wollastonita que possui apenas a granulometria < 0,30 mm. Os tratamentos foram aplicados em três solos: Latossolo Vermelho distrófico, Nitossolo Vermelho eutrófico e Neossolo Quartzarênico. No 1° experimento aplicou-se quatro doses (1, 2, 4 e 8 t ha-1) afim de obter a curva de neutralização dos solos, para determinar a dose a ser aplicada de cada fração granulométrica para atingir o pH 5,5. O 2° experimento constou da aplicação das doses ecomendadas no primeiro experimento e com o valor de pH obtido aos 90 dias de incubação calculou-se a eficiência relativa de reatividade para cada fração granulométrica, tomando-se como referência o pH obtido com a aplicação do calcário dolomítico na peneira... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract : The efficiency and quality of materials for soil acidity correction are based on the reactivity rate (RR) and neutralization power (NP), and it is indicated by the effective calcium carbonate (ECC). Considering that there are few studies establishing reactivity rates of silicate particle-size fractions, lime values are used as reference. Three experiments were carried in the period July/2008 to October/2009 out in greenhouse in College of Agricultural Sciences, UNESP, Botucatu - SP, aiming to compare reactivity rates of silicate particle-size fractions in three soils with distinct textures cultivated with alfalfa plants. Slag, calcium silicate, calcium+magnesium silicate, wollastonite, dolomitic and calcitic lime were the materials evaluated. The first and second experiments consisted of four particlesize fractions obtained from each material, separated by the sieves no. 10 (Ø = 2 mm), no. 20 (Ø = 0.84 mm), no. 50 (Ø = 0.30 mm) and bottom (Ø < 0.30 mm), except that the wollastonite has the particle-size fraction < 0,30 mm. Treatments were applied in three soils: Oxisol, Hapludox and Quartzipsamment. In the first study, four doses were applied (1, 2, 4 and 8 t ha- 1) to obtain soil neutralization curve and thus determine which one was to be applied for each particle-size fraction to reach pH 5.5. The second study consisted of the application of the recommended doses from the first experiment. Then, with the pH values obtained 90 days after incubation, the relative efficiency of reactivity was calculated for each particle-size fraction based on the pH of dolomitic lime from the last sieve. Afterwards, RR and ECC were calculated. In the third study, the doses to increase soil base saturation up to 80% were calculated based on both the PRNT reported in the laboratory and in the experiment. Treatments were incubated... (Complete abstract click electronic access below) / Mestre

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / O presente trabalho analisa os africanos livres durante a construção da Casa de Correção da Corte. Para conhecer esse caminho, a investigação começa com a lei de 7 de novembro de 1831 e termina com a inauguração da primeira prisão com trabalho do país, em 1850. A dissertação tem como fio condutor o estudo da produção jurídica sobre os africanos livres e para isso, levam-se em consideração três eixos documentais: as leis brasileiras, os relatórios ministeriais e os despachos burocráticos relativos à administração da casa. O exame dessa material possibilita a operacionalização da noção de disciplinarização e como ela construiu a identidade dos africanos livres como trabalhadores. A disciplina como instrumento de formação de trabalhadores foi o meio de viabilizar a utilização dessa mão de obra. Esse estudo, ao problematizar aos africanos livres na condição de trabalhadores debate as formas de trabalho no Império do Brasil, colocando em pauta as experiências da escravidão e as expectativas do trabalho livre. E nessa análise, o sentido político das enunciações jurídicas foi o cerne para entendimentos desses agentes históricos. Dessa forma, eles foram instruídos e capacitados em diferentes ofícios, na qual recebiam salário e pagavam por sua subsistência. Nesse sentido, a dissertação se estrutura na relação da produção jurídica com a experimentação do trabalho livre. / The present work analyzes the free africans during the construction of the House of Correction. To know this pathway, the investigation begins with the November 7, 1831 law and ends with the inauguration of the countrys first working prison, in 1850. The dissertations conduction wire is the study of the legal production regarding the free africans and, for this, three documental axes are taken into consideration: the Brazilian laws, the ministerial reports and the bureaucratic orders regarding the house administration. The examination of this material allows the operationalization of the notion of discipline undertaken by Michel Foucault. The discipline as an instrument of workers formation was the means to enable the use of the free africans working power. This study, by discussing the free africans as workers, debates the ways of work in Brazilian Empire, putting in evidence the experiences of slavery and the expectations of free labor. And in this discussion, the political sense of legal assertions was the core for the understanding of them as free workers. This way, they were instructed and trained in different trades, in which they received salaries and paid for their own subsistence. At this point, the dissertation structures itself in the relation between legal production and free labor experimentation.