Att manuellt välja en eller flera meningar ur en filmrecension att använda som citat kan vara en tidskrävande uppgift. Denna rapport utvärderar övervakade maskininlärningsmodeller för att skapa en prototyp som automatiskt kan välja lämpliga citat ur recensioner. Utifrån resultatet av en litteraturstudie valdes två modeller att implementera och utvärdera på data bestående av filmrecensioner och tillhörande manuellt valda citat. Av arbetets två implementerade modeller, BERT med Inter-Sentence Transformer och BERT med ett artificiellt neuronnät, visade den sistnämnda marginellt bättre resultat. Modellerna utvärderades med ROUGE och jämfördes med tidigare studiers toppresultat inom automatisk textsummering. Slutsatsen är att de modeller som utvärderades inte presterar tillräckligt väl inom problemområdet för att motivera en driftsättning utan ytterligare utvecklingsarbete. Dock visar resultaten att det finns potential i att de utvärderade tillvägagångssätten delvis kan ersätta manuella val av citat i framtiden. / To choose a number of sentences from a movie review to use as a quote can be time consuming if done manually. This thesis evaluates supervised machine learning models to create a prototype that automatically can choose such quotes. The thesis chose, based on a literature study, two models to implement and evaluate on data consisting of movie reviews and their respective corresponding manually chosen quotes. Out of the thesis two implemented models, BERT with Inter-Sentence Transformer and BERT with an artificial neural network, the latter showed marginally better results. The models were evaluated with ROUGE and was compared with state-of-the-art models regarding automatic text summarization. The conclusion is that the models that were evaluated do not perform well enough for the problem to motivate full deployment without further development efforts. However, the results show that there is potential that the evaluated methods can partially replace manual labour when choosing quotes.

La documentation des programmes aide les développeurs à mieux comprendre le code source pendant les tâches de maintenance. Toutefois, la documentation n’est pas toujours disponible ou elle peut être de mauvaise qualité. Le recours à la redocumentation s’avère ainsi nécessaire. Dans ce contexte, nous proposons de faire la redocumentation en générant des commentaires par application de techniques de résumé par extraction. Pour mener à bien cette tâche, nous avons commencé par faire une étude empirique pour étudier les aspects quantitatifs et qualitatifs des commentaires. En particulier, nous nous sommes intéressés à l’étude de la distribution des commentaires par rapport aux différents types d’instructions et à la fréquence de documentation de chaque type. Aussi, nous avons proposé une taxonomie de commentaires pour classer les commentaires selon leur contenu et leur qualité. Suite aux résultats de l’étude empirique, nous avons décidé de résumer les classes Java par extraction des commentaires des méthodes/constructeurs. Nous avons défini plusieurs heuristiques pour déterminer les commentaires les plus pertinents à l’extraction. Ensuite, nous avons appliqué ces heuristiques sur les classes Java de trois projets pour en générer les résumés. Enfin, nous avons comparé les résumés produits (les commentaires produits) à des résumés références (les commentaires originaux) en utilisant la métrique ROUGE. / Programs documentation is very useful to programmers during maintenance tasks, especially for program comprehension. However, the documentation is not always available or it may be badly written. In such cases, redocumentation becomes so necessary. In this work, we propose a redocumentation technique that consists in generating comments by using extraction summary techniques. For this purpose, we conducted an empirical study to analyze the quantitave and qualitative aspects of comments. Particularly, we were interested in studying comment distribution over different types of construct and studying the frequency of documentation for each construct type. We propose a comment taxonomy to classify them according to their content and quality. Given the results of the empirical study, we decided to summarize Java classes by extracting the comments of methods and constructors. We defined several heuristics in order to determine the most relevant comments to be extracted. After that, we applied these heuristics to Java classes from three projects in order to generate summaries. Finally, we compared the generated summaries (generated comments) to the reference ones (original comments) by using the metric ROUGE.

Processos estocásticos complexos são muitas vezes utilizados em modelagem, com o intuito de capturar uma maior proporção das principais características dos sistemas biológicos. A descrição do comportamento desses sistemas tem sido realizada por muitos amostradores baseados na distribuição a posteriori de Monte Carlo. Modelos probabilísticos que descrevem esses processos podem levar a funções de verossimilhança computacionalmente intratáveis, impossibilitando a utilização de métodos de inferência estatística clássicos e os baseados em amostragem por meio de MCMC. A Computação Bayesiana Aproximada (ABC) é considerada um novo método de inferência com base em estatísticas de resumo, ou seja, valores calculados a partir do conjunto de dados (média, moda, variância, etc.). Essa metodologia combina muitas das vantagens da eficiência computacional de processos baseados em estatísticas de resumo com inferência estatística bayesiana uma vez que, funciona bem para pequenas amostras e possibilita incorporar informações passadas em um parâmetro e formar uma priori para análise futura. Nesse trabalho foi realizada uma comparação entre os métodos de estimação, clássico, bayesiano e ABC, para estudos de simulação de modelos simples e para análise de dados de dinâmica populacional. Foram implementadas no software R as distâncias modular e do máximo como alternativas de função distância a serem utilizadas no ABC, além do algoritmo ABC de rejeição para equações diferenciais estocásticas. Foi proposto sua utilização para a resolução de problemas envolvendo modelos de interação populacional. Os estudos de simulação mostraram melhores resultados quando utilizadas as distâncias euclidianas e do máximo juntamente com distribuições a priori informativas. Para os sistemas dinâmicos, a estimação por meio do ABC apresentou resultados mais próximos dos verdadeiros bem como menores discrepâncias, podendo assim ser utilizado como um método alternativo de estimação. / Complex stochastic processes are often used in modeling in order to capture a greater proportion of the main features of natural systems. The description of the behavior of these systems has been made by many Monte Carlo based samplers of the posterior distribution. Probabilistic models describing these processes can lead to computationally intractable likelihood functions, precluding the use of classical statistical inference methods and those based on sampling by MCMC. The Approxi- mate Bayesian Computation (ABC) is considered a new method for inference based on summary statistics, that is, calculated values from the data set (mean, mode, variance, etc.). This methodology combines many of the advantages of computatio- nal efficiency of processes based on summary statistics with the Bayesian statistical inference since, it works well for small samples and it makes possible to incorporate past information in a parameter and form a prior distribution for future analysis. In this work a comparison between, classical, Bayesian and ABC, estimation methods was made for simulation studies considering simple models and for data analysis of population dynamics. It was implemented in the R software the modular and maxi- mum as alternative distances function to be used in the ABC, besides the rejection ABC algorithm for stochastic differential equations. It was proposed to use it to solve problems involving models of population interaction. The simulation studies showed better results when using the Euclidean and maximum distances together with informative prior distributions. For the dynamic systems, the ABC estimation presented results closer to the real ones as well as smaller discrepancies and could thus be used as an alternative estimation method.

O escopo deste trabalho é analisar a tutela jurisdicional diferenciada, suas caracteríticas, limitações e utilizá-la como uma das soluções para a morosidade do Poder Judiciário. Para isso, investigaremos o próprio conceito de tutela jurisdicional, fazendo uma análise histórica e chegando ao conceito atual. Após, estudaremos cada um dos tipos específicos de tutela jurisdicional diferenciada, examinando suas peculiaridades e problemática. Nesse capítulo estudaremos a tutela cautelar, antecipada e a de evidência (grande novidade incluída no projeto do novo Código de Processo Civil). Também estudaremos outras medidas de cognição sumária, apesar de esses outros tipos não apresentarem qualquer problema para se estabilizarem. No capítulo 3 trataremos das características específicas da tutela jurisdicional diferenciada, principalmente da sumariedade na cognição, requisito essencial para esse tipo de tutela. Em seguida, serão examinados os princípios constitucionais-processuais, como o devido processo legal e o princípio da igualdade e de como achar um equilíbrio entre a eficiência exigida pelo Estado e a duração razoável do processo, dever do Poder Judiciário e garantia dos jurisdicionados. A questão da coisa julgada, da sucumbência e da fundamentação da decisão serão trabalhadas no capítulo 5. Pretendemos demonstrar nesse capítulo de como é possível outorgar força de coisa julgada para as decisões de cognição sumária, e não apenas a preclusão endoprocessual como prevista no projeto de novo Código de Processo Civil. Por fim, analisamos as propostas legislativas já apresentadas pelo Instituto Brasileiro de Direito Processual e o projeto de novo Código de Processo Civil. Ao longo do trabalho foram analisados a doutrina nacional e estudos de outros países, investigados a aceitação desses institutos e a recepção pelo Poder Judiciário, pelos advogados e pelos jurisdicionados. / The scope of this work is to analyze the non-ordinary court protection, their characters, and limitations and to find a way to use it as one of the solutions to the slow pace of the Judiciary branch. For this, we investigate the concept of court protection, making an historical analysis up to the present concept. After this, we study each of the specific types of non-ordinary court protection, examining their peculiarities and their problems. In this chapter we study provisional proceedings, evidence court protection (a new institute included in the project Code of Civil Procedure). We also study other non-ordinary proceedings. In chapter 3 we study the specific characteristics of the non-ordinary court protection, mostly the summary cognition, an essential requirement. Furthermore, we analyze the constitutional procedural principles, like due process of law, the principle of equality; and find a balance between efficiency and reasonable length of the proceedings, which is duty of the Judiciary branch and a guarantee of the population. We also analyze the res judicata, loss of suit and opinion in chapter 5. In this chapter, we want to demonstrate the possibility of granting res judicata to non-ordinary court protection and not only to the preclusion of the process. Ultimately, we analyze the legal proposes already presented by Instituto Brasileiro de Direito Processual and the project of a new Code of Civil Procedure. In this work, we also study national doctrine and foreign legal writing. We investigate these new institutes, mostly the evidence court protection and the reception by the Judiciary branch, both from lawyers and the population.

Made available in DSpace on 2016-04-28T19:33:51Z (GMT). No. of bitstreams: 1 Tharsila Dantas Prates.pdf: 3487333 bytes, checksum: 0704472d812a2abbfb50ccad492e63bd (MD5) Previous issue date: 2014-09-02 / The present work is in the line of research Reading, writing and teaching of Portuguese. The theme is the use of the microblogging service Twitter for writing s activities to grade school students. The aim of this work is to propose activities using Twitter for Portuguese classes. The specific objectives are: describe the Twitter and propose writing s activities that permit a concise writing and that prepare the students to summarization of texts. The theoretical base is the research in Textual Linguistics at the social, cognitive and interactional perspective, used for define textual coherence (BEAUGRANDE & DRESSLER, 1981; BEAUGRANDE, 1997; KOCH & TRAVAGLIA, 2011, 2012); show the standards of textuality (BEAUGRANDE & DRESSLER, 1981; KOCH & TRAVAGLIA, 2011, 2012); the macrostructure or the principal idea of text (VAN DIJK, 2008) and the roles to reduce semantic information (VAN DIJK & KINTSCH, 1975). In the presence of access to new technologies and the need of teaching of concise and coherent writing, we developed a proposal with the use of Twitter to be applied to an opinion article sort of text worked by educators in nine year of grade school. We believe that microblogging can be an aid to the Portuguese teacher make students develop the power of synthesis and writing coherently / Este trabalho insere-se na linha de pesquisa Leitura, escrita e ensino de Língua Portuguesa e tem como tema a utilização do microblog Twitter em atividades de escrita voltadas a alunos do 9º ano do Ensino Fundamental. O objetivo geral é apresentar uma proposta de utilização do Twitter nas aulas de Língua Portuguesa. Os objetivos específicos são: identificar recursos que o microblog Twitter oferece e propor atividades que permitam ao aluno o exercício da síntese e a preparação para a escrita de resumo. Para desenvolvermos a proposta, embasamo-nos na perspectiva sociocognitivo-interacional da Linguística Textual no que diz respeito a aspectos da coerência textual, tais como: sua definição (BEAUGRANDE & DRESSLER, 1981; BEAUGRANDE, 1997; KOCH & TRAVAGLIA, 2011, 2012); os fatores que contribuem para essa coerência (BEAUGRANDE & DRESSLER, 1981; KOCH & TRAVAGLIA, 2011, 2012); a identificação da macroestrutura ou ideia principal do texto (VAN DIJK, 2008) e as regras de redução da informação semântica (VAN DIJK & KINTSCH, 1975). Diante da possibilidade de acesso às novas tecnologias e da necessidade do ensino da escrita coerente e concisa, desenvolvemos uma proposta com o uso do Twitter, para ser aplicada a um artigo de opinião, gênero trabalhado no último ano do Ensino Fundamental na Rede Municipal de São Paulo. Acreditamos que o microblog pode constituir um auxílio ao professor de Língua Portuguesa para fazer com que o educando desenvolva o poder de síntese e a escrita com coerência

Submitted by Filipe dos Santos (fsantos@pucsp.br) on 2017-08-17T13:21:00Z No. of bitstreams: 1 Otávio Brasil de Moraes.pdf: 5548770 bytes, checksum: f76d20e7f7854b63714dd8e1030f214f (MD5) / Made available in DSpace on 2017-08-17T13:21:00Z (GMT). No. of bitstreams: 1 Otávio Brasil de Moraes.pdf: 5548770 bytes, checksum: f76d20e7f7854b63714dd8e1030f214f (MD5) Previous issue date: 2017-08-07 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / Fundação de Amparo à Pesquisa do Estado do Amazonas - FAPEAM / In this dissertation, our general objective is to propose the usage of the notions of microstructures and macrostructures in the production of summaries of opinion articles. Throughout the research, developed together with students of a secondary school in the city of Manaus, we have highlighted significant differences between summaries produced in the “traditional” manner and those produced according to our proposal. Our theoretical framework is Textual Linguistics of text grammars, especially the proposal by Van Dijk (1996). We have also considered the concept of text developed in the 1970’s and 80’s by Van Dijk and Kinstch. Moreover, we have highlighted more recent studies, such as Marquesi (2004), Delphino (1991) and Machado (2004). Methodologically, the research was developed as follows: initially, in the first class, we requested that students produced an abstract of an opinion article following the traditional perspective in teaching how to produce summaries, in which it is generally emphasized only the selection of the main ideas in a text. In the second class, we discussed the concepts of microstructures and macrostructures for the production of summaries, and, in the third class, we requested that students produced a second summary using those notions. We then analyzed twenty-eight summaries produced by 14 students, which allowed us to verify positive contributions from the proposal of working with textual macrostructures. The data analysis chapter contains as examples 6 summaries chosen among the 28 that were analyzed, of which 3 (three) were written with the traditional perspective and the other 3 (three) were written in the framework of microstructures and macrostructures. After, we discuss the results, which clearly point out the contributions of the studies of textual microstructures and macrostructures to the production of summaries / Nesta dissertação, temos como objetivo geral propor a utilização das noções de micro e macroestruturas para a produção de resumos de artigos de opinião. Ao longo da pesquisa, desenvolvida junto aos discentes do ensino médio de uma escola estadual da cidade de Manaus, destacamos diferenças significativas entre os resumos produzidos de forma “tradicional” e aqueles produzidos segundo a nossa proposta. Para tanto, tomamos como base teórica a Linguística Textual das gramáticas de texto, sobretudo a proposta de van Dijk (1996). Consideramos também o conceito de texto desenvolvido durante as décadas de 1970 e 1980 por van Dijk e Kinstch. Ademais, destacamos trabalhos mais recentes que tratam da produção de resumos de textos segundo a perspectiva desses estudos, tais como Marquesi (2004), Leite (2006), Delphino (1991) e Machado (2004). Metodologicamente, a pesquisa se desenvolveu da seguinte forma: inicialmente, na primeira aula, solicitamos aos alunos que produzissem um resumo de um artigo de opinião, obedecendo à perspectiva tradicional de ensino de produção de resumo, na qual de modo geral se enfatiza apenas a identificação das ideias principais do texto; na segunda aula, trabalhamos os conceitos de micro e macroestruturas para a produção de textos-resumo e, na terceira aula, solicitamos aos discentes que produzissem um segundo resumo, no caso considerando a utilização das noções de micro e macroestrutura. Após essa etapa, analisamos vinte e oito resumos produzidos por 14 alunos, o que nos permitiu verificar os pontos positivos da proposta que considera o trabalho com as macroestruturas textuais. No capítulo de análise, apresentamos a título de exemplificação a análise de 6 (seis) resumos escolhidos entre os 28 analisados, sendo 3 (três) elaborados segundo a perspectiva tradicional de ensino de produção de resumo e 3 (três) elaborados pelo viés micro/ macroestrutura. Em seguida, discutimos os resultados que sinalizaram claramente as contribuições da perspectiva que considera os estudos de micro e macroestrutura textual para a produção de resumos

Regional development in Canada is uneven. However, it is less clear whether native development in Canada is similarly patterned as non-native development. As well, it is not evident whether variations exist in the differences between the level of development of native and non-native people in different regions of Canada.<p> Development is a multi-dimensional process involving economic and social change. The primary purpose of this thesis is to examine the use of the indicator approach to create a measure of development which encompasses both the economic and social dimensions of development. The secondary purposes of the thesis are: (1) to use the indicator approach to compute a measure of development to identify spatial variations in the level of development of native people in Canada in 1981, and (2) to use development indicators to examine the spatial variations in the concept of dualism as it applies to native and non-native residents of Canada.<p> The data source for this analysis is the 1981 Native Summary Tape, a special tape produced by Census Canada containing a multitude of socio-economic data for natives and non-natives at the national, provincial, census division and census subdivision level. The primary statistical technique used in the computation of the development measures is principal components analysis. The development measures are computed at the census division level and mapping aids in the identification of the spatial patterns.<p> The results from this analysis suggest that the indicator approach to measuring development presented in this thesis is a valuable tool in the identification of spatial variations in development. This is an important procedure for researchers who wish to examine development in its socio-economic totality. Native development and dualism between native and non-native Canadians do vary spatially. The general pattern is higher native development and lower dualism between natives and non-natives in the larger urban centers and in central Canada.

No description available.

Regional development in Canada is uneven. However, it is less clear whether native development in Canada is similarly patterned as non-native development. As well, it is not evident whether variations exist in the differences between the level of development of native and non-native people in different regions of Canada.<p> Development is a multi-dimensional process involving economic and social change. The primary purpose of this thesis is to examine the use of the indicator approach to create a measure of development which encompasses both the economic and social dimensions of development. The secondary purposes of the thesis are: (1) to use the indicator approach to compute a measure of development to identify spatial variations in the level of development of native people in Canada in 1981, and (2) to use development indicators to examine the spatial variations in the concept of dualism as it applies to native and non-native residents of Canada.<p> The data source for this analysis is the 1981 Native Summary Tape, a special tape produced by Census Canada containing a multitude of socio-economic data for natives and non-natives at the national, provincial, census division and census subdivision level. The primary statistical technique used in the computation of the development measures is principal components analysis. The development measures are computed at the census division level and mapping aids in the identification of the spatial patterns.<p> The results from this analysis suggest that the indicator approach to measuring development presented in this thesis is a valuable tool in the identification of spatial variations in development. This is an important procedure for researchers who wish to examine development in its socio-economic totality. Native development and dualism between native and non-native Canadians do vary spatially. The general pattern is higher native development and lower dualism between natives and non-natives in the larger urban centers and in central Canada.

In this study, meta-analysis of diagnostic tests, Summary Receiver Operating Characteristic (SROC) curve, bivariate random effects and Hierarchical Summary Receiver Operating Characteristic (HSROC) curve theories have been discussed and accuracy in literature of Fine Needle Aspiration (FNA) biopsy that is used in the diagnosis of masses in breast cancer (malignant or benign) has been analyzed. FNA Cytological (FNAC) examination in breast tumor is, easy, effective, effortless, and does not require special training for clinicians. Because of the uncertainty related to FNAC&lsquo / s accurate usage in publications, 25 FNAC studies have been gathered in the meta-analysis. In the plotting of the summary ROC curve, the logit difference and sums of the true positive rates and the false positive rates included in the meta-analysis&lsquo / s codes have been generated by SAS. The formula of the bivariate random effects model and hierarchical summary ROC curve is presented in context with the literature. Then bivariate random effects implementation with the new SAS PROC GLIMMIX is generated. Moreover, HSROC implementation is generated by SAS PROC HSROC NLMIXED. Curves are plotted with RevMan Version 5 (2008). It has been stated that the meta-analytic results of bivariate random effects are nearly identical to the results from the HSROC approach. The results achieved through both random effects meta-analytic methods prove that FNA Cytology is a diagnostic test with a high level of distinguish over breast tumor.