Zhan Siyan. / "November 2004." / Thesis (Ph.D.)--Chinese University of Hong Kong, 2004. / Includes bibliographical references (p. 133-152) / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Mode of access: World Wide Web. / Abstracts in English and Chinese.

Les travaux de cette thèse présentent le développement d'algorithmes de classification de documents d'une part, ou d'analyse de réseaux complexes d'autre part, en s'appuyant sur la prétopologie, une théorie qui modélise le concept de proximité. Le premier travail développe un cadre pour la classification de documents en combinant une approche de topicmodeling et la prétopologie. Notre contribution propose d'utiliser des distributions de sujets extraites à partir d'un traitement topic-modeling comme entrées pour des méthodes de classification. Dans cette approche, nous avons étudié deux aspects : déterminer une distance adaptée entre documents en étudiant la pertinence des mesures probabilistes et des mesures vectorielles, et effet réaliser des regroupements selon plusieurs critères en utilisant une pseudo-distance définie à partir de la prétopologie. Le deuxième travail introduit un cadre général de modélisation des Réseaux Complexes en développant une reformulation de la prétopologie stochastique, il propose également un modèle prétopologique de cascade d'informations comme modèle général de diffusion. De plus, nous avons proposé un modèle agent, Textual-ABM, pour analyser des réseaux complexes dynamiques associés à des informations textuelles en utilisant un modèle auteur-sujet et nous avons introduit le Textual-Homo-IC, un modèle de cascade indépendant de la ressemblance, dans lequel l'homophilie est fondée sur du contenu textuel obtenu par un topic-model. / The work of this thesis presents the development of algorithms for document classification on the one hand, or complex network analysis on the other hand, based on pretopology, a theory that models the concept of proximity. The first work develops a framework for document clustering by combining Topic Modeling and Pretopology. Our contribution proposes using topic distributions extracted from topic modeling treatment as input for classification methods. In this approach, we investigated two aspects: determine an appropriate distance between documents by studying the relevance of Probabilistic-Based and Vector-Based Measurements and effect groupings according to several criteria using a pseudo-distance defined from pretopology. The second work introduces a general framework for modeling Complex Networks by developing a reformulation of stochastic pretopology and proposes Pretopology Cascade Model as a general model for information diffusion. In addition, we proposed an agent-based model, Textual-ABM, to analyze complex dynamic networks associated with textual information using author-topic model and introduced Textual-Homo-IC, an independent cascade model of the resemblance, in which homophily is measured based on textual content obtained by utilizing Topic Modeling.

Thesis (M.S.P.H.)--University of South Florida, 2008. / Title from PDF of title page. Document formatted into pages; contains 26 pages. Includes bibliographical references.

Thesis (Ph.D. in Biostatistics) -- University of Colorado at Denver and Health Sciences Center, 2006. / Typescript. Includes bibliographical references (leaves 72-75). Free to UCD Anschutz Medical Campus. Online version available via ProQuest Digital Dissertations;

Thesis (Ph. D.)--University of Alabama at Birmingham, 2008. / Title from first page of PDF file (viewed Sept. 23, 2008). Includes bibliographical references.

Thesis (Ph.D. in Biostatistics, Department of Preventive Medicine and Biometrics) -- University of Colorado Denver, 2007. / Typescript. Includes bibliographical references (leaves 89-93). Free to UCD affiliates. Online version available via ProQuest Digital Dissertations;

Thesis (M.Sc.)--Memorial University of Newfoundland, Faculty of Medicine, 1999. / Typescript. Bibliography: leaves 82-110.

Supervised machine learning models require a labeled data set of high quality in order to perform well. Available text data often exists in abundance, but it is usually not labeled. Labeling text data is a time consuming process, especially in the case where multiple labels can be assigned to a single text document. The purpose of this thesis was to make the labeling process of clinical reports as effective and effortless as possible by evaluating different multi-label active learning strategies. The goal of the strategies was to reduce the number of labeled documents a model needs, and increase the quality of those documents. With the strategies, an accuracy of 89% was achieved with 2500 reports, compared to 85% with random sampling. In addition to this, 85% accuracy could be reached after labeling 975 reports, compared to 1700 reports with random sampling.

Made available in DSpace on 2014-12-17T15:07:09Z (GMT). No. of bitstreams: 1 FranciscoJCS_DISSERT.pdf: 3148033 bytes, checksum: 1d6e571d5bcd941b37b01b5a7eb6cb07 (MD5) Previous issue date: 2014-02-24 / Coordena??o de Aperfei?oamento de Pessoal de N?vel Superior / This study is part of interactional perspective, focusing on Conversation Analysis theories, from the Textual Interactive Perspective and Text Linguistics . This research, from its guiding questions, aims at understanding the interaction between teacher and students in the process of the knowledge construction as well as at describing, analyzing and understanding aspects of topical organization speech in the classroom in elementary school, observing the opening and closing procedures of the speech topics in that particular space. Considering that the procedures for opening and closing of discursive topics occur through language marks, we tried to identify which speech marks are used in the opening and closing of the topics studied in the classroom, in interaction during the collaborative process of the discourse established between teacher and students. Therefore, this study is based on authors who analyze specific questions of the text in real context of language use: Koch (1993, 1999), Jubran et al (1991), Jubran (2006), Pine (2005), Penhavel (2010), Galembeck (2012), Barros (1991), Marcuschi (1986 , 1990, 1991 , 1998 , 1999, 2003 , 2004a), Kerbrat - Orecchioni (2006), Favero (1999, 2002) and Galv?o (2004, 2010). As a methodology of investigation, the study is focused on the postulates of ethnographic research in order to carry out data collection, through audio and video recordings which were transcribed, according to the NURC project proposal, with some adaptations. Data analysis showed that the procedures for opening and closing of the speech topics occurred by the use of discourse markers, in particular the marker "then", allowing us to understand that these elements are important in the topical organization speech, contributing to ensure textual cohesion and coherence. We conclude that the organization of the discursive topic in the classroom occurs through events that support the explicitness of the content of teaching and learning, considering the diverse necessity of an institutional academic plan, whose main objective is the construction of knowledge / Este trabalho se insere em perspectivas interacionais, com base nos pressupostos da An?lise da Conversa??o, da Perspectiva Textual Interativa e da Lingu?stica de Texto. A partir da quest?o norteadora que busca uma compreens?o acerca da intera??o entre professor e alunos no processo de constru??o do conhecimento, temos como objetivos: descrever, analisar e interpretar aspectos da organiza??o t?pica em sala de aula no Ensino Fundamental, observando os procedimentos de abertura e fechamento dos t?picos nesse espa?o espec?fico. Partindo da expectativa de que os procedimentos de abertura e fechamento dos t?picos discursivos ocorrem por meio de marcas lingu?sticas, buscamos identificar que marcas s?o usadas nas aberturas e fechamentos dos t?picos na sala de aula investigada, no transcurso das intera??es durante o processo colaborativo do discurso instaurado entre professora e alunos. Assim sendo, embasamo-nos em autores que analisam quest?es espec?ficas do texto em situa??o concreta de uso da linguagem, entre estes, Koch (1993, 1999), Jubran et al (1991), Jubran (2006), Pinheiro (2005), Penhavel (2010), Galembeck (2012), Barros (1991), Marcuschi (1986, 1990, 1991, 1998, 1999, 2003, 2004a), Kerbrat-Orecchioni (2006), F?vero (1999, 2002) e Galv?o (2004, 2010). Metodologicamente, orientamo-nos nos postulados da pesquisa etnogr?fica, a fim de realizar a coleta dos dados, atrav?s de grava??es em ?udio e v?deo, que foram transcritos, em seguida, conforme a proposta do projeto NURC, com algumas adapta??es. A an?lise dos dados ressaltados revelou que os procedimentos de abertura e de fechamento dos t?picos ocorreram pelo uso de marcadores discursivos, em especial o marcador ent?o , propiciando-nos ? compreens?o de que esses elementos s?o importantes na organiza??o t?pica, contribuindo para assegurar a coer?ncia e a coes?o textual. Conclu?mos que a organiza??o do t?pico discursivo em sala de aula acontece mediante ocorr?ncias que auxiliam a explicita??o do conte?do de ensino e aprendizagem, tendo em vista ?s diversas necessidades de um plano acad?mico institucional, cujo objetivo principal ? a constru??o do conhecimento

Um dos fenômenos mais investigados na literatura sintática recente sobre o português brasileiro (PB), dentro do quadro teórico gerativista e também de outras teorias linguísticas, é aquele que se refere ao que viria a ficar conhecido como tópico-sujeito. Este trabalho se debruçou sobre este fenômeno sintático, mais especificamente, o tópico-sujeito locativo, um tipo de tópico-sujeito em que um constituinte com interpretação locativa figura em posição pré-verbal. Os aspectos investigados mais de perto referem-se à grade argumental e à estrutura eventiva dos verbos envolvidos, bem como ao modo de licenciamento dos dois DPs participantes da estrutura eventiva que expressa a relação locativa. Depois de uma revisão da literatura relevante sobre o tópico-sujeito e as bases do gerativismo, a hipótese derivacional que assumimos neste trabalho considerou uma estrutura semelhante à das sentenças transitivas sem, no entanto, a projeção de Voice agentivo. Postulamos, em seu lugar, um núcleo aplicativo alto, com traços de pessoa, que identifica um evento apto a receber uma locação e introduz um novo participante afetado, com uma relação temática locativa entre ele e o VP. Do ponto de vista diacrônico, supomos que o parâmetro envolvido no aparecimento do tópico locativo na gramática do PB estaria na mesma rede paramétrica relacionada à mudança no parâmetro do sujeito nulo. Foi feita também uma pesquisa com corpora históricos abrangendo o período dos séculos XIX ao XXI, com o objetivo de rastrear um possível desenvolvimento das construções com sintagmas locativos em posição pré-verbal, em variação com os locativos em posição pós-verbal, na expectativa de poder encontrar ocorrências de construções que apresentem as propriedades que identifiquem o tópico-sujeito locativo. Foram utilizados anúncios, cartas pessoais e entrevistas, todos circunscritos ao estado de São Paulo. A metodologia utilizada foi quantitativa e buscou contabilizar variáveis presentes em todas as sentenças com sintagmas de interpretação locativa. / syntactic literature, inside the theoretical frame of the Generative Syntax program and also another linguistic theories, is the one that has come to be known as topic-subject. This work focused on this syntactic phenomenon, more specifically the locative topic, a subtype of topic-subject in which a constituent with locative interpretation appears in a pre-verbal position. The more closely investigated aspects refer to the argument structure and event structure of the verbs involved, as well as to the licensing mode of the two DPs participating in the event structure that expresses the locative relation. After a literature revision about the topic-subject and the foundations of the generative enterprise, the derivational hypothesis that we assume in this work considers a structure similar to that of the transitive sentences without, however, the agentive Voice projection. We consider instead a high applicative head, with person features, whose function is to identify an event capable of receiving a location and to introduce a new affected participant, with a locative thematic relation between it and the VP. From the diachronic point of view, we assume that the parameter involved in the implementation of the locative topic in the BP grammar would be in the same parametric network related to the change in the null subject parameter. Besides, a research was made with historical corpora spanning the period from the nineteenth century to the twenty-first century, aiming to track a probable development of the constructions with locative phrases in pre-verbal position, in variation with locatives in post-verbal position, waiting to find occurrences of constructions showing the topic-subject properties. We used ads, personal letters and interviews, all of which were limited to the state of São Paulo. The methodology used was quantitative and sought to account for variables present in all sentences with phrases associated to locative interpretation.