Zugl.: Univ. of Queensland, Diss., 2005

La variabilité en langue est une source majeure de difficultés dans la plupart des applications du traitement automatique des langues. Elle se manifeste dans le fait qu'une même idée ou un même événement peut être exprimé avec des mots ou des groupes de mots différents ayant la même signification dans leur contexte respectif. Capturer automatiquement des équivalences sémantiques entre des unités de texte est une tâche complexe mais qui s'avère indispensable dans de nombreux contextes. L'acquisition a priori de listes d'équivalences met à disposition des ressources utiles pour, par exemple, améliorer le repérage d'une réponse à une question, autoriser des formulations différentes en évaluation de la traduction automatique, ou encore aider des auteurs à trouver des formulations plus adaptées. Dans cette thèse, nous proposons une étude détaillée de la tâche d'acquisition de paraphrases sous-phrastiques à partir de paires d'énoncés sémantiquement liés. Nous démontrons empiriquement que les corpus parallèles monolingues, bien qu'extrêmement rares, constituent le type de ressource le plus adapté pour ce genre d'étude. Nos expériences mettent en jeu cinq techniques d'acquisition, représentatives de différentes approches et connaissances, en anglais et en français. Afin d'améliorer la performance en acquisition, nous réalisons la combinaison des paraphrases produites par ces techniques par une validation reposant sur un classifieur automatique à maximum d'entropie bi-classe. Un résultat important de notre étude est l'identification de paraphrases qui défient actuellement les techniques étudiées, lesquelles sont classées et quantifiées en anglais et français. Nous examinons également dans cette thèse l'impact de la langue, du type du corpus et la comparabilité des paires des énoncés utilisés sur la tâche d'acquisition de paraphrases sous- phrastiques. Nous présentons le résultat d'une analyse de la performance des différentes méthodes testées en fonction des difficultés d'alignement des paires de paraphrases d'énoncés. Nous donnons, ensuite, un compte rendu descriptif et quantitatif des caractéristiques des paraphrases trouvées dans les différents types de corpus étudiés ainsi que celles qui défient les approches actuelles d'identification automatique.

Automatic paraphrase discovery is an important task in natural language processing. Many systems use paraphrases for improve performance e.g. systems for question answering, information retrieval or document summarization. In this thesis, we explain basic concepts e.g. paraphrase or paraphrase pattern. Next we propose some methods for paraphrase discovery from various resources. Subsequently we propose an unsupervised method for discovering paraphrase from large plain text based on context and keywords between NE pairs. In the end we explain evaluation metods in paraphrase discovery area and then we evaluate our system and compare it with similar systems.

Given a sentence, a paraphrase generation system produces a sentence that says the same thing but usually in a different way. The paraphrase generation problem can be formulated in the machine translation paradigm; instead of translation of English to a foreign language, the system translates an English sentence (for example) to another English sentence. Quirk et al. (2004) demonstrated this approach to generate almost 90% acceptable paraphrases. However, most of the sentences had little variation from the original input sentence. Leveraging syntactic information, this thesis project presents an approach that successfully generated more varied paraphrase sentences than the approach of Quirk et al. while maintaining coverage of the proportion of acceptable paraphrases generated. The ParaMeTer system (Paraphrasing by MT) identifies syntactic chunks in paraphrase sentences and substitutes labels for those chunks. This enables the system to generalize movements that are more syntactically plausible, as syntactic chunks generally capture sets of words that can change order in the sentence without losing grammaticality. ParaMeTer then uses statistical phrase-based MT techniques to learn alignments for the words and chunk labels alike. The baseline system followed the same pattern as the Quirk et al. system - a statistical phrase-based MT system. Human judgments showed that the syntactic approach and baseline both achieve approximately the same ratio of fluent, acceptable paraphrase sentences per fluent sentences. These judgments also showed that the ParaMeTer system has more phrase rearrangement than the baseline system. Though the baseline has more within-phrase alteration, future modifications such as a chunk-only translation model should improve ParaMeTer's variation for phrase alteration as well.

Made available in DSpace on 2014-06-11T19:26:50Z (GMT). No. of bitstreams: 0 Previous issue date: 2008-03-03Bitstream added on 2014-06-13T18:07:50Z : No. of bitstreams: 1 biasottoholmo_m_me_arafcl.pdf: 428789 bytes, checksum: f0b3e2d5887e93e46b29319c5ab63567 (MD5) / Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Este trabalho visa à comprovação da hipótese de que os aprendizes de uma língua estrangeira, ao traduzirem um enunciado, partem de um esquema chamado léxis, para então anexar a ele as categorias gramaticais de modalidade, determinação, aspecto e diátese, por meio de marcas lingüísticas, e que, ao passo em que o aprendizado evolui, há uma melhor compreensão dessas categorias. Fundamentada na Teoria das Operações Predicativas e Enunciativas, de Antoine Culioli, esta pesquisa tem por objetivo apresentar reflexões que contribuam com o desenvolvimento das abordagens de ensino/aprendizado de língua estrangeira. Assim, propomos uma abordagem enunciativa da língua que leve em consideração os saberes epilingüístico e metalingüístico dos aprendizes, os processos de parafrasagem e a construção das categorias gramaticais. / This work has in view the confirmation of the hypothesis that learners of a foreign language, when translating an utterance, start from a framework called léxis, so that they can annex to it the grammatical categories of modality, determination, aspect and diathesis, through linguistic markers, and that while the learning evolve, there is a better understanding of these categories. Based on the Theory of Predicative and Enunciative Operations, by Antoine Culioli, this study aims to present reflections that can contribute to the development of foreign language teaching/learning approaches. Thus, we propose an enunciative approach of language which takes into consideration learner’s epilinguistic and metalinguistic knowledge, the paraphrase processes and the construction of grammatical categories.

L’objectif de notre thèse est de décrire les états linguistiques chez des enfants bilingues de 10 ans habitant à Paris, à travers leurs reformulations en français et en japonais de l’histoire « Tom et Julie ». Cette histoire comporte plusieurs éléments qui permettront d’analyser l’acquisition linguistique. Nous examinerons premièrement la « complexité » des reformulations. Deuxièmement, en examinant les procédures de reformulation, nous catégoriserons six formes de reformulations. Cette classification nous permettra de saisir les tendances des reformulations chez les enfants bilingues. Troisièmement, nous ferons l’étude comparée des reformulations en français et en japonais des deux textes. En dernier lieu, nous étudierons individuellement les reformulations en français et en japonais chez les enfants bilingues. Comment les enfants manipulent-ils les deux langues, qui sont si éloignées l’une de l’autre, pour exprimer la même histoire ? Quelle performance linguistique les enfants nous montrent-ils ? Peut-on trouver des règles linguistiques qui permettent aux enfants bilingues d’apprendre efficacement les deux langues ? Notre étude tente d’expliquer l’état linguistique des enfants bilingues en répondant à ces questions. / The goal of our thesis is to describe the linguistic states of bilingual children who are 10 years old and live in Paris, through their reformulation in French and in Japanese of the story « Tom et Julie ». This story contains some elements that permit us to analyze linguistic acquisition. Firstly, we examine the complexity of reformulations. Secondly, in examining the procedures of reformulation, we categorize six forms of reformulation. This classification will permit us to catch the tendencies of reformulations in the case of bilingual children. Thirdly, we compare French reformulation and Japanese reformulation in the two texts. Lastly, we will study these reformulations individually. How children manipulate French and Japanese that are so distant from each other, to express the same story ? What linguistic performance do they show ? Can we find linguistic rules which permit bilingual children to learn efficiently two languages ? Our study tries to explain the linguistic state of bilingual children in responding to these questions.

Le Liber Prefigurationum Christi et Ecclesie est un poème anonyme latin datable de la fin du XIème siècle. Il n’est présent en entier que dans un seul manuscrit parisien. Il s’agit d’une réécriture de la Bible comportant 2670 hexamètres dactyliques. L’auteur a paraphrasé un certain nombre d’épisodes de l’Ancien Testament et a ajouté des commentaires exégétiques et moraux. La thèse comprend, dans une première partie, une introduction au texte et à sa traduction. Elle étudie la structure du poème, les techniques de réécriture, l’utilisation des commentaires exégétiques et des poètes antérieurs à la rédaction. Elle compare le Liber avec d’autres réécritures bibliques tardo-antiques ou médiévales. Elle dresse enfin un portrait de l’anonyme en insistant sur sa culture, ses techniques de versification, ses structures de pensée et ses intentions. La deuxième partie de l’ouvrage comprend une nouvelle édition critique du texte ainsi que sa première traduction. Elle s’accompagne d’un apparat des gloses et des sources bibliques. La troisième partie du travail est constituée d’un commentaire « épisode par épisode » de l’ensemble du poème. Il évoque la place de chaque passage dans l’économie de l’œuvre (en particulier dans les cas d’antéposition ou de postposition par rapport à la matière biblique), la réécriture de l’hypotexte biblique, la présence et la fonction des commentaires, les sources patristiques ainsi que la justification de nos choix d’édition lorsqu’ils diffèrent du texte établi par G. Dinkova en 2007. / The Liber Prefigurationum Christi et Ecclesie is an anonymous poem, written in Latin, which can be dated from the end of the eleventh century. The unabridged version is to be found in only one Parisian manuscript. The verse composition is a bible rewriting containing 2670 hexameters. The author has versified many episodes of the Old Testament and has added exegetical and moral commentaries. The thesis contains, in its first part, an introduction to the Latin text and its translation, which studies the poem structure, the rewriting modes, the exploitation of the exegetical commentaries and of the poets who have written prior to the Liber composition. It also compares the Liber with other biblical rewritings either from late Antiquity or from the Middle Ages. Finally it draws the portrait of the anonymous poet by emphasizing his culture, his versification, his way of thinking and his aims. The second part of the work comprises a new critical edition of the text as well as its first translation. It also includes an apparatus of the glosses and of the biblical references. The third part is a commentary “episode by episode” of the whole poem which mentions the place of each passage in the work (particularly in case of anteposition or postposition), the rewriting of the biblical hypotext, the presence and function of the commentaries, the doctores used by the author, as well as the explanation of our edition choices, when they are different from the text written by G. Dinkova in 2007.

Mémoire numérisé par la Direction des bibliothèques de l'Université de Montréal.

Este estudo tem por objetivo investigar o procedimento parafrástico em redações de vestibular com base na hipótese de que nos textos analisados os distintos tipos de procedimentos parafrásticos empregados permitem vislumbrar diferentes concepções de língua no ensino da escrita. Desse modo, as reformulações do tipo paráfrase discursiva produzidas pelos escreventes, a partir dos posicionamentos históricos presentes nos textos da coletânea, constituem um mecanismo dialógicoargumentativo. O trabalho se fundamenta no conceito de paráfrase proposto pela corrente teórica da Análise do Discurso de linha francesa. De forma específica objetiva-se: a) investigar a ocorrência das chamadas paráfrase linguística e paráfrase discursiva em redações de vestibular em que a proposta de escrita é direcionada pelos textos da coletânea; b) analisar a paráfrase discursiva tomando-a como um procedimento dialógico-argumentativo no texto dissertativo; c) apontar razões para as ocorrências de reformulações parafrásticas encontradas. A presente investigação partiu de uma amostra de 205 redações das quais 137 apresentaram ao menos uma paráfrase de uma das filiações sócio-históricas veiculadas nos textos da coletânea fornecida na prova de redação do vestibular da FUVEST/2006. Das 137 redações que exibiram formulações em relação de paráfrase com os textos da coletânea, 77 delas apresentaram paráfrases do tipo linguística e 60 paráfrases do tipo discursiva. Os resultados obtidos mostraram que, de fato, na produção do texto dissertativo a reformulação parafrástica do tipo paráfrase discursiva funciona como um mecanismo argumentativo na articulação das formações discursivas veiculadas pelos escreventes. Nessas redações verificou-se que a paráfrase discursiva atualiza os discursos dos textos da coletânea nas novas condições de produção. Por sua vez, constatou-se que a reformulação parafrástica do tipo paráfrase linguística constitui um elemento que compõe o quadro das condições de produção da escrita escolar, o que tomamos como um indicador de que, nesse tipo de procedimento parafrástico, a concepção normativa da língua orientou a maioria dos escreventes / This paper aims to investigate paraphrastic procedures in university application essays based upon the hypothesis that distinct types of paraphrastic procedures allow discerning different conceptions of language. Therefore, the discoursive paraphrastic reformulation produced by candidates, as a consequence of historical positions displayed in a collection of texts provided as required writing starters, constitutes a dialogical/argumentative process. The study adopts the concept of paraphrase proposed by the French School of Discourse Analysis. Specifically, the study aims to: a) investigate the occurrence of linguistic and discoursive paraphrases in admissions essays in which the writing proposal is directed by a collection of texts; b) analyze the discoursive paraphrase as a dialogical-argumentative procedure; c) indicate reasons of the presence of found paraphrases. The present investigation initiated with a set of 205 essays, from which 137 exhibited at least one paraphrase of one historical filiation introduced by the collection of texts provided in the examination essay of FUVEST (University of São Paulos entry exam) in the year of 2006. The results obtained demonstrate that in the essay writing the discoursive paraphrase in fact works as an argumentative process in the articulation of the discourse formations presented by the candidates. In these essays it was verified that the paraphrastic process modifies the source discourses in the new conditions of the discoursive productions. On the other hand, the results suggest that the paraphrastic reformulation of the linguistic type constitutes an element that forms the conditions of language production in the school, which we consider an indicator that this type of paraphrase derived from the prescriptive conception of language rules the majority of the applicants writings. From the 137 essays that exhibited paraphrases of the collection of texts, 77 displayed paraphrases of the linguistic type and 60 paraphrases of the discoursive type

Este estudo tem por objetivo investigar o procedimento parafrástico em redações de vestibular com base na hipótese de que nos textos analisados os distintos tipos de procedimentos parafrásticos empregados permitem vislumbrar diferentes concepções de língua no ensino da escrita. Desse modo, as reformulações do tipo paráfrase discursiva produzidas pelos escreventes, a partir dos posicionamentos históricos presentes nos textos da coletânea, constituem um mecanismo dialógicoargumentativo. O trabalho se fundamenta no conceito de paráfrase proposto pela corrente teórica da Análise do Discurso de linha francesa. De forma específica objetiva-se: a) investigar a ocorrência das chamadas paráfrase linguística e paráfrase discursiva em redações de vestibular em que a proposta de escrita é direcionada pelos textos da coletânea; b) analisar a paráfrase discursiva tomando-a como um procedimento dialógico-argumentativo no texto dissertativo; c) apontar razões para as ocorrências de reformulações parafrásticas encontradas. A presente investigação partiu de uma amostra de 205 redações das quais 137 apresentaram ao menos uma paráfrase de uma das filiações sócio-históricas veiculadas nos textos da coletânea fornecida na prova de redação do vestibular da FUVEST/2006. Das 137 redações que exibiram formulações em relação de paráfrase com os textos da coletânea, 77 delas apresentaram paráfrases do tipo linguística e 60 paráfrases do tipo discursiva. Os resultados obtidos mostraram que, de fato, na produção do texto dissertativo a reformulação parafrástica do tipo paráfrase discursiva funciona como um mecanismo argumentativo na articulação das formações discursivas veiculadas pelos escreventes. Nessas redações verificou-se que a paráfrase discursiva atualiza os discursos dos textos da coletânea nas novas condições de produção. Por sua vez, constatou-se que a reformulação parafrástica do tipo paráfrase linguística constitui um elemento que compõe o quadro das condições de produção da escrita escolar, o que tomamos como um indicador de que, nesse tipo de procedimento parafrástico, a concepção normativa da língua orientou a maioria dos escreventes / This paper aims to investigate paraphrastic procedures in university application essays based upon the hypothesis that distinct types of paraphrastic procedures allow discerning different conceptions of language. Therefore, the discoursive paraphrastic reformulation produced by candidates, as a consequence of historical positions displayed in a collection of texts provided as required writing starters, constitutes a dialogical/argumentative process. The study adopts the concept of paraphrase proposed by the French School of Discourse Analysis. Specifically, the study aims to: a) investigate the occurrence of linguistic and discoursive paraphrases in admissions essays in which the writing proposal is directed by a collection of texts; b) analyze the discoursive paraphrase as a dialogical-argumentative procedure; c) indicate reasons of the presence of found paraphrases. The present investigation initiated with a set of 205 essays, from which 137 exhibited at least one paraphrase of one historical filiation introduced by the collection of texts provided in the examination essay of FUVEST (University of São Paulos entry exam) in the year of 2006. The results obtained demonstrate that in the essay writing the discoursive paraphrase in fact works as an argumentative process in the articulation of the discourse formations presented by the candidates. In these essays it was verified that the paraphrastic process modifies the source discourses in the new conditions of the discoursive productions. On the other hand, the results suggest that the paraphrastic reformulation of the linguistic type constitutes an element that forms the conditions of language production in the school, which we consider an indicator that this type of paraphrase derived from the prescriptive conception of language rules the majority of the applicants writings. From the 137 essays that exhibited paraphrases of the collection of texts, 77 displayed paraphrases of the linguistic type and 60 paraphrases of the discoursive type