Cette thèse se place dans le cadre de services web en dépassant leur description pour considérer leur structuration en réseaux (réseaux d'interaction et réseaux de similitude). Nous proposons des méthodes basées sur les motifs, la modélisation probabiliste et l'analyse des concepts formels, pour améliorer la qualité des services découverts. Trois contributions sont alors proposées: découverte de services diversifiés, recommandation de services et cohérence des communautés de services détectées. Nous structurons d'abord les services sous forme de réseaux. Afin de diversifier les résultats de la découverte, nous proposons une méthode probabiliste qui se base à la fois sur la pertinence, la diversité et la densité des services. Dans le cas de requêtes complexes, nous exploitons le réseau d'interaction de services construit et la notion de diversité dans les graphes pour identifier les services web qui sont susceptibles d'être composables. Nous proposons également un système de recommandation hybride basé sur le contenu et le filtrage collaboratif. L'originalité de la méthode proposée vient de la combinaison des modèles thématiques et les motifs fréquents pour capturer la sémantique commune maximale d'un ensemble de services. Enfin, au lieu de ne traiter que des services individuels, nous considérons aussi un ensemble de services regroupés sous forme de communautés de services pour la recommandation. Nous proposons dans ce contexte, une méthode qui combine la sémantique et la topologie dans les réseaux afin d'évaluer la qualité et la cohérence sémantique des communautés détectées, et classer également les algorithmes de détection de communautés. / As a part of this thesis, we exceed the description of web services to consider their structure as networks (i.e. similarity and interaction web service networks). We propose methods based on patterns, topic models and formal concept analysis, to improve the quality of discovered services. Three contributions are then proposed: (1) diversified services discovery, (2) services recommendation and (3) consistency of detected communities. Firstly, we propose modeling the space of web services through networks. To discover the diversified services corresponding to a given query, we propose a probabilistic method to diversify the discovery results based on relevancy, diversity and service density. In case of complex requests, it is necessary to combine multiple web services to fulfill this kind of requests. In this regard, we use the interaction web service network and the diversity notion in graphs to identify all possible services compositions. We also propose a new hybrid recommendation system based on both content and collaborative filtering. Its originality comes from the combination of probabilistic topic models and pattern mining to capture the maximal common semantic of a set of services. Finally, instead of processing individual services, we consider a set of services grouped into service communities for the recommendation. We propose in this context, a new method combining both topology and semantics to evaluate the quality and the semantic consistency of detected communities, and also rank the detection communities algorithms.

[pt] O estudo do sujeito é de extrema importância na prática pedagógica do ensino da Língua Portuguesa. Muitas questões têm sido levantadas sobre a impropriedade com que as definições se aplicam aos exemplos e exercícios das gramáticas escolares. Nessa dissertação apresentamos uma revisão dessas definições, estabelecendo comparações com os exemplos dados nas gramáticas tradicionais, materiais didáticos e frases que surgem no uso não-padrão da língua. Nosso arcabouço teórico é pautado na abordagem funcionalista, na qual priorizamos a gramática de valências que coloca o sujeito gramatical como um complemento e a gramática sistêmica de M. A K. Halliday que apresenta o sujeito desempenhando papéis definidos pelos verbos. A partir das questões levantadas e da análise das definições dadas temos a convicção de que aquilo que a prática tradicional convencionou como modelo para o ensino de análise sintática precisa ser reavaliado, uma vez que verificamos severas contradições entre a teoria e prática. / [en] The study of the subject is of distinct importance for the pedagogical practice of the Portuguese language teaching. Many issues have been raised about the inadequacy of the definitions applied to the examples and exercises in grammars. In this dissertation, we present a review of these definitions, stablishing comparisons among the examples given in traditional grammars, teaching materials and sentences that appear in the non-pattern use of the language. Our theoretical framework is based on the functionalist approach, in which we have chosen the valency grammar which classifies the grammatical subject as a complement and the systemic grammar of M.A. K. Halliday which presents the subject performing papers defined by the verbs. Based on the issues raised and based on the analysis of the definitions we are convinced that that model which the traditional practice has established as the model for the teaching of syntatic analysis must be re-evaluated, since we have noticed many serious contradictions between the theory and the practice.

Audiobook reading has traditionally been used for educational purposes but has in recent times grown into a popular alternative to the more traditional means of consuming literature. In order to differentiate themselves from other players in the market, but also provide their users enjoyable literature, several audiobook companies have lately directed their efforts on producing own content. Creating highly rated content is, however, no easy task and one reoccurring challenge is how to make a bestselling story. In an attempt to identify latent features shared by successful audiobooks and evaluate proposed methods for literary quantification, this thesis employs an array of frameworks from the field of Statistics, Machine Learning and Natural Language Processing on data and literature provided by Storytel - Sweden’s largest audiobook company. We analyze and identify important features from a collection of 3077 Swedish books concerning their promotional and literary success. By considering features from the aspects Metadata, Theme, Plot, Style and Readability, we found that popular books are typically published as a book series, cover 1-3 central topics, write about, e.g., daughter-mother relationships and human closeness but that they also hold, on average, a higher proportion of verbs and a lower degree of short words. Despite successfully identifying these, but also other factors, we recognized that none of our models predicted “bestseller” adequately and that future work may desire to study additional factors, employ other models or even use different metrics to define and measure popularity. From our evaluation of the literary quantification methods, namely topic modeling and narrative approximation, we found that these methods are, in general, suitable for Swedish texts but that they require further improvement and experimentation to be successfully deployed for Swedish literature. For topic modeling, we recognized that the sole use of nouns provided more interpretable topics and that the inclusion of character names tended to pollute the topics. We also identified and discussed the possible problem of word inflections when modeling topics for more morphologically complex languages, and that additional preprocessing treatments such as word lemmatization or post-training text normalization may improve the quality and interpretability of topics. For the narrative approximation, we discovered that the method currently suffers from three shortcomings: (1) unreliable sentence segmentation, (2) unsatisfactory dictionary-based sentiment analysis and (3) the possible loss of sentiment information induced by translations. Despite only examining a handful of literary work, we further found that books written initially in Swedish had narratives that were more cross-language consistent compared to books written in English and then translated to Swedish.

Tratar grandes quantidades de dados é uma exigência dos modernos algoritmos de mineração de texto. Para algumas aplicações, documentos são constantemente publicados, o que demanda alto custo de armazenamento em longo prazo. Então, é necessário criar métodos de fácil adaptação para uma abordagem que considere documentos em fluxo, e que analise os dados em apenas um passo sem requerer alto custo de armazenamento. Outra exigência é a de que essa abordagem possa explorar heurísticas a fim de melhorar a qualidade dos resultados. Diversos modelos para a extração automática das informações latentes de uma coleção de documentos foram propostas na literatura, dentre eles destacando-se os modelos probabilísticos de tópicos. Modelos probabilísticos de tópicos apresentaram bons resultados práticos, sendo estendidos para diversos modelos com diversos tipos de informações inclusas. Entretanto, descrever corretamente esses modelos, derivá-los e em seguida obter o apropriado algoritmo de inferência são tarefas difíceis, exigindo um tratamento matemático rigoroso para as descrições das operações efetuadas no processo de descoberta das dimensões latentes. Assim, para a elaboração de um método simples e eficiente para resolver o problema da descoberta das dimensões latentes, é necessário uma apropriada representação dos dados. A hipótese desta tese é a de que, usando a representação de documentos em grafos bipartidos, é possível endereçar problemas de aprendizado de máquinas, para a descoberta de padrões latentes em relações entre objetos, por exemplo nas relações entre documentos e palavras, de forma simples e intuitiva. Para validar essa hipótese, foi desenvolvido um arcabouço baseado no algoritmo de propagação de rótulos utilizando a representação em grafos bipartidos. O arcabouço, denominado PBG (Propagation in Bipartite Graph), foi aplicado inicialmente para o contexto não supervisionado, considerando uma coleção estática de documentos. Em seguida, foi proposta uma versão semissupervisionada, que considera uma pequena quantidade de documentos rotulados para a tarefa de classificação transdutiva. E por fim, foi aplicado no contexto dinâmico, onde se considerou fluxo de documentos textuais. Análises comparativas foram realizadas, sendo que os resultados indicaram que o PBG é uma alternativa viável e competitiva para tarefas nos contextos não supervisionado e semissupervisionado. / Handling large amounts of data is a requirement for modern text mining algorithms. For some applications, documents are published constantly, which demand a high cost for long-term storage. So it is necessary easily adaptable methods for an approach that considers documents flow, and be capable of analyzing the data in one step without requiring the high cost of storage. Another requirement is that this approach can exploit heuristics in order to improve the quality of results. Several models for automatic extraction of latent information in a collection of documents have been proposed in the literature, among them probabilistic topic models are prominent. Probabilistic topic models achieve good practical results, and have been extended to several models with different types of information included. However, properly describe these models, derive them, and then get appropriate inference algorithms are difficult tasks, requiring a rigorous mathematical treatment for descriptions of operations performed in the latent dimensions discovery process. Thus, for the development of a simple and efficient method to tackle the problem of latent dimensions discovery, a proper representation of the data is required. The hypothesis of this thesis is that by using bipartite graph for representation of textual data one can address the task of latent patterns discovery, present in the relationships between documents and words, in a simple and intuitive way. For validation of this hypothesis, we have developed a framework based on label propagation algorithm using the bipartite graph representation. The framework, called PBG (Propagation in Bipartite Graph) was initially applied to the unsupervised context for a static collection of documents. Then a semi-supervised version was proposed which need only a small amount of labeled documents to the transductive classification task. Finally, it was applied in the dynamic context in which flow of textual data was considered. Comparative analyzes were performed, and the results indicated that the PBG is a viable and competitive alternative for tasks in the unsupervised and semi-supervised contexts.

O presente trabalho é um estudo da utilidade que o Cinema oferece como recurso na Educação Médica inserido num contexto humanístico, na sistemática acadêmica da Medicina de Família. O autor desenvolve, na primeira parte da obra, uma extensa fundamentação da pesquisa através de uma reflexão sobre as diversas realidades que rodeiam o médico que, sendo também educador, propõe-se ensinar atitudes e valores humanísticos aos alunos. O itinerário reflexivo percorre as várias situações onde se descreve o encontro com o Paciente, com o Aluno, com a Medicina de Família e, finalmente, com o Humanismo, para chegar até o Cinema como manifestação particular das humanidades e possível recurso educacional. As extensas referências bibliográficas, às que se somam as experiências biográficas e motivações do autor, levantam questionamentos e interrogações que fundamentam e justificam ume pesquisa de ordem filosófico e humanísta, como a apresentada neste trabalho. Sublinha-se, neste ponto, que o Humanismo deve ser para o médico fonte de conhecimentos, uma verdadeira ferramenta de trabalho. O corpo da pesquisa segue uma abordagem de natureza qualitativa perspectiva que é também explicada e justificada na obra- e tem como objetivo avaliar a possível utilidade do Cinema como recurso educacional. O cenário da pesquisa instala-se em diversos contextos educacionais, utilizando vídeos editados com cenas de filmes comerciais que são projetados a grupos de alunos, acompanhados de comentários do professor, seguindo-se uma discussão do grupo participante. A análise de resultados revela várias categorias de temáticas emergentes, constando-se, em primeiro lugar, como através desta experiência é possível criar um espaço apropriado para a discussão das expectativas, dilemas e motivações do aluno, permitindo uma compreensão do universo no qual o estudante está imerso. Nota-se também que o contato com o cinema facilita a expressão afetiva do aluno, inserido como está numa cultura que prioriza a emoção, a imagem concreta, as histórias de vida e os exemplos pontuais que decorrem de situações determinadas. Através desta experiência eduacional instala-se uma linguagem de comunicação onde o professor pode transmitir valores e fomentar atitudes, e o aluno consegue representar as realidades vividas, simbolizando-as no cinema. Deste modo, a experiência com o cinema propicia trabalhar as motivações do estudante a partir da sua dimensão afetiva, fomentando atitudes e promovendo nele o hábito da reflexão, que são a base de uma educação humanista e elementos essenciais para o seu processo de formação como médico e como pessoa. / The purpose of this academic work is to study the value of using cinema (movies), set in the context of Family Medicine, as a humanistic resource in medical education. In the first part, the author grounds the research by reflecting on the several realities that surround physician-educators who want to teach students about attitudes and humanistic values. The reflexive itinerary of this work addresses several sets of relationships - here called encounters - by describing the encounter with the patient, the encounter with the students, the encounter with the discipline of family medicine, and lastly the encounter with humanism, finally reaching cinema in particular as a possible educational resource. In each encounter the authors summarizes his own experiences using a reflective approach, enlarged upon them using the bibliography, and develops several questions and inquiries for justifying a research project such as this one, displayed from a philosophical and humanistic perspective. Humanism as a source of acquaintance and as a daily work tool for doctors is stressed at this point. The research design is utilizes a qualitative perspective, which is also justified, and attempts to appraise the usefulness of movies as educational resource. The research scenario includes several educational contexts in which movies scenes are edited as clips and shown to the students along with the facultys commentary, followed by discussions among the group. Results analysis discloses various categories of emerging subjects. First, this experience works to create room in which students expectations, dilemmas and motivation can be discussed, more comprehensively addressing the universe in which students are immersed. Exposing students to this cinema experience facilitates their affective expression, since they are raised in a culture which prioritizes emotions, images, life stories and role models coming from specific situations. A real means of communication is established and through it faculty can teach about values and foster attitudes, and the students are able to represent their own reality, symbolized for them in the movies realities. The educational cinema experience herein related helps students to deal with the aspects of their motivation that emerges from the affective domain, is useful in promoting new attitudes and reflection, and nurturing essential features related to their development process as doctors and human beings.

Cette thèse analyse la transition topicale en anglais américain à l’aide d’un corpus audio de conversations spontanées entre proches. L’objet d’étude principal est l’action interactionnelle qui consiste à changer de topique discursif, ainsi que les diverses stratégies linguistiques que les participants ont à leur disposition. Trois modalités de marquage sont prises en compte : les questions, les marqueurs de discours, et le registre de la voix. Chaque modalité est analysée pour sa contribution individuelle, ainsi que pour les associations avec d’autres modalités qu’elle peut occasionner. Se pencher sur différentes modalités de marquage crée une vue d’ensemble composite de l’influence que la trajectoire topicale d’une conversation a sur sa grammaire et sa prosodie. Dans le cadre d’une approche mixte mêlant analyses qualitatives et quantitatives, cette étude se situe à la croisée de plusieurs cadres théoriques, empruntant tant à l’analyse conversationnelle et à la linguistique interactionnelle pour l’analyse qualitative située, qu’à la linguistique de corpus de par ses méthodes quantitatives telles que le codage systématique des données et le recours aux statistiques. Ce projet multi-domaines est complété par une comparaison entre conversations typiques et atypiques. Les personnes schizophrènes peuvent connaître des difficultés dans la gestion des topiques d’une conversation, ce qui peut occasionner des transitions non-canoniques. Comparer ce type de données à celles de participants typiques apporte un éclairage supplémentaire sur certaines des attentes, préférences et standards, par ailleurs moins visibles lorsque la transition topicale est plus aisée. / The research presented in this dissertation analyzes topic transition in American English interaction, focusing on audio recordings of spontaneous conversations between friends and relatives. The main object of inquiry is the interactional action of transitioning to a new discourse topic, as well as the different linguistic strategies that participants have at their disposal. Three main types of cues are investigated: questions, discourse markers, and pitch register. Each type of cue is analyzed for its individual contribution to topic transition design, as well as for the way it can combine with, supplement, or contradict other cues. Analyzing different types of cues – verbal and prosodic – creates a composite picture of the various ways in which the topic trajectory of a conversation shapes its grammar – including its prosody. This study uses a mixed-methods approach which draws on the qualitative-oriented theoretical frameworks of Conversation Analysis and Interactional Linguistics, combining them with quantitative methods used in Corpus Linguistics, such as systematic coding and statistics. This multi-domain account is completed by elaborating a comparison between typical and atypical interactions. Persons suffering from schizophrenia can experience difficulties in managing the topics of a conversation, and they can produce non-canonical transitions. Comparing their data with that of typical participants thus sheds light on some of the expectations, preferences and standard formats which can otherwise remain hidden when topic transition goes smoothly.

Orientador: Helio Amante Miot / Resumo: Introdução: Minoxidil tópico é o único medicamento com aprovação pelo FDA (Food and Drug Administration) para tratamento da alopecia de padrão feminino (APF). Muitas pacientes interrompem o tratamento prematuramente devido a falta de eficácia, intolerância ou por alteração na textura dos fios de cabelo. Minoxidil oral mostrou-se efetivo para tratamento da APF em estudo não controlado. Objetivo: Comparar eficácia, segurança e tolerabilidade do minoxidil 1 mg oral uma vez ao dia versus minoxidil 5% solução tópica uma vez ao dia no tratamento da APF. Métodos: Estudo prospectivo, randomizado, paralelo, comparativo, avaliador cego com duração de 24 semanas conduzido em um único centro de janeiro de 2017 a março de 2018 incluindo 52 mulheres (18-65 anos) com APF. Participantes foram randomizadas para receber minoxidil oral 1 mg ao dia ou minoxidil 5% tópico uma vez ao dia. O desfecho primário foi a variação na densidade de fios na área alvo. Desfechos secundários: variação na densidade de fios terminais na área alvo, escore na avaliação fotográfica panorâmica por avaliadores cegos, variação no escore da escala de queda, variação no escore do Women's Androgenetic Alopecia Quality of Life Questionnaire (WAA-QoL). Resultados: Participantes de ambos os grupos apresentaram melhora na densidade de fios na área alvo (p<0,01), porém, sem diferença entre os grupos (p=0,09): oral 12% (IC95%: 8,0 – 16,1%) e tópico 7,2% (IC95%: 1,5 - 12,9%). Houve melhora na densidade de fios terminais: oral 6... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: Introduction: Topical minoxidil is the only FDA (Food and Drug Administration) approved drug for female pattern hair loss (FPHL). Many patients discontinue treatment prematurely due to lack of efficacy, intolerance or altered hair texture. Oral minoxidil was effective for FPHL in uncontrolled studies. Objectives: To compare the efficacy, safety and tolerability of once-daily treatment with 1 mg oral minoxidil versus once-daily 5% minoxidil solution applied topically for FPHL. Methods: A 24-week, prospective, randomized, open-label, parallel, two-arm comparative, evaluator-blinded study conducted in a single center from January 2017 through March 2018 including 52 women (18-65 years old) with FPHL. Participants were randomly assigned to receive once daily minoxidil 1 mg orally or once a day minoxidil 5% solution applied topically. The primary endpoint was change from baseline in hair density from a target area at week 24. Secondary endpoints were change in terminal hair density, global photographic assessment by three group-blinded evaluators, hair shedding score, and the Women's Androgenetic Alopecia Quality of Life Questionnaire (WAA-QoL). Results: Participants of both groups had improvement of hair density in the target area (p<0.01), without difference between the groups (p=0.09): oral 12% (CI95%: 8.0- 16.1%) and topical 7,2% (CI95%: 1.5-12.9%). There was improvement on terminal hair density: oral 6% (IC95%: 2.9 – 8.6%) and topical 2,7% (IC95%: -1.4 – 6.8%), with no differ... (Complete abstract click electronic access below) / Doutor

This study addresses questions of gender and genre in early writing by drawing on systemic linguistic theory, It is a longitudinal case study that compares the writing development of two children, a boy and a girl/ who learned to write in classrooms that adopted an approach to writing known in Australia as 'process writing1, The children's written texts were analysed using the systemic functional grammar as developed by MAK, Hallidey and the models of genre and register as proposed by J,R, Martin. The children were followed for the first two and a half years of their schooling, from the first day of kindergarten to the middle of grade two. They were observed weekly during the daily ‘writing time’ and all texts were collected. Although the children were ostensibly 'free’ to determine both the writing topics and text types they produced, systemic analysis revealed that: 1) the majority of texts written were of one genre, the Observation genre, in which the children reconstructed their personal experience with family and friends and offered an evaluation of it. 2) a significant pattern of gender differences occurred within this genre, such that the boy reconstructed experience in terms of the male cultural stereotype of being an active participant in the world, while the girl reconstructed experience in terms of the female stereotype of being a more passive observer of experience. It is the strength of systemic linguistic analysis that it revealed how the choices the children made in language were constrained by a number of social and cultural contexts, including: a) the teacher's theoretical orientation to literacy; b) the models of spoken and written language available to the children; and c) the ideology of gender in the culture. In particular, the analysis made visible how children appropriate the meanings of their culture and socialise themselves into gender roles by constructing the ideology of gender in their writing. The study contributes to an understanding of genres by offering a revised description of the Observation genre, which derives from the Observation Comment genre originally identified by Martin and Rothery (1981). It also raises a number of implications for teacher training and classroom practice, including the need for: 1) increased teacher consciousness about gender and genre, especially an understanding that choices in language are socially constructed 2) a critical reassessment of the notion of 'free topic choice’ promoted by 'process writing' pedagogy, a practice which may limit choice and tacitly support the gender status quo.

本研究旨在探討文章結構與主題知識對國小學童說明文寫作表現的影響。 正式實驗的受試包括台北縣秀朗國小五年級三個班級共 138名學童，分別 接受文章結構明示、暗示與不提示三種實驗處理，然後受試均各寫一篇對 主題知識熟悉與不熟悉的說明文，依變項為組織結構與內容觀念兩個寫作 表現的成績。研究結果發現，在扣除智力與國語科成績的影響後，在寫作 組織結構的表現上，提示文章結構並沒有幫助，反而是以不提示文章結構 的控制組表現最優；在內容觀念向度的表現上，以暗示文章結構的方式最 有助於寫作表現，而且文章結構的提示方式與寫作主題知識對內容觀念產 出也具有交互作用的效果，當寫作主題熟悉時，暗示文章結構最能提高內 容觀念向度的寫作表現。無論是在內容觀念或組織結構，都是以當寫作者 對寫作主題知識熟悉時的表現優於當寫作者對寫作主題知識不熟悉時，因 此本研究建議，在寫作教學上，教師應該針對學生所熟悉的事物來命題， 同時與其他學科的領域知識相結合，以豐富寫作內容。

This thesis is based on a conversation about gender identity and the pronoun hen in an episode of the radio programme Tendens on Swedish public radio, P1. The participants of the conversation are a scholar and a journalist/show-host. The goal of the research has been to analyse how the conversation is constructed from its context and which topics that occur. The goal of the research has also been to analyse how the participants of the conversation position themselves and how the conversation composes and re-composes norms and values regarding gender identity in society. Critical discourse analysis, completed by Conversation analysis has been the theoretical and methodological approach. The result shows that the conversation functions well within its context and that the host Söderqvist and scholar Milles undertakedifferent parts and contributions in the interaction. The host undertakes the part as the presenter and questioner of problems and topics, which Milles then refute in her role as an expert. The result also shows that Söderqvist and Milles give voice to a diversity of topical and interactional positionings in the conversation, but that Söderqvist mainly positions himself as the audience’s representative. Furthermore the result shows that the participants recreate gender identity norms in the conversation. Milles dictums also carry tendencies to create and broaden norms.