The problem considered in this thesis is the development of an autonomous prototype robot capable of gathering sensory information from its environment allowing it to provide feedback on the condition of specific targets to aid in maintenance of hydro equipment. The context for the solution to this problem is based on the power grid environment operated by the local hydro utility. The intent is to monitor power line structures by travelling along skywire located at the top of towers, providing a view of everything beneath it including, for example, insulators, conductors, and towers. The contribution of this thesis is a novel robot design with the potential to prevent hazardous situations and the use of rough coverage feedback modified reinforcement learning algorithms to establish behaviours.

The problem considered in this thesis is the development of an autonomous prototype robot capable of gathering sensory information from its environment allowing it to provide feedback on the condition of specific targets to aid in maintenance of hydro equipment. The context for the solution to this problem is based on the power grid environment operated by the local hydro utility. The intent is to monitor power line structures by travelling along skywire located at the top of towers, providing a view of everything beneath it including, for example, insulators, conductors, and towers. The contribution of this thesis is a novel robot design with the potential to prevent hazardous situations and the use of rough coverage feedback modified reinforcement learning algorithms to establish behaviours.

In this study, Turkey&rsquo / s Internet visibility will be analyzed based on data to be collected from multiple different resources (such as / Google, Yahoo, Altavista, Bing and AOL). Analysis work will involve inspection of DNS queries, Web crawling and some other similar techniques. Our goal is to investigate global Internet and find webs that has common pattern of representing Internet visibility of Turkey and compare their characteristics with other webs&#039 / on the world and discover their similarities and differences.

The evolution, development and function of the embryonic and larval peripheral nervous system (PNS) were investigated in a number of dipteran species, including Aedes aegypti, Hermetia pallescens, Lucilia cuprina, Bactrocera tryoni and Drosophila melanogaster. Comparative immuno-cytochemical data was generated for developmental proteins in the embryonic PNS and degenerative PCR was employed to identify homologous proneural genes amongst the species. Immunocytochemistry revealed aspects of sense-organ evolution amongst the Diptera, particularly an increase in sense-cell number and number of sensilla comprising proprioceptive organs in the crawling versus swimming larvae. The function of putative proprioceptive cells was investigated via mutant analysis and laser ablation. Results indicate that the multiple-dendrite (md) and bipolar-dendrite (bd) neurons contribute in a cumulative and roughly equal fashion to maintain peristaltic waves in crawling larvae. This work highlights the usefulness of the dipteran PNS as a model of adaptive evolution that can be investigated via developmental mechanisms.

A successful immune response depends on prompt and sufficient recruitment of leukocytes from the circulation to infected or injured sites. Mobilization of leukocytes to hypoxic tissues is vital for angiogenesis, i.e. the formation of new blood vessels from preexisting vasculature, and thus crucial for tissue growth and regeneration. Deviations from normal leukocyte recruitment drive a variety of pathologies, including chronic inflammation, autoimmune diseases and cancer, for which therapeutic options are limited or unspecific. Understanding the mechanisms by which the body controls leukocyte recruitment is therefore critical for the development of novel therapeutic strategies. The present investigations focused on delineating the mechanisms behind leukocyte mobilization from the bloodstream to afflicted sites, by means of in vivo imaging techniques and in vitro assays. We demonstrate that, in response to inflammation, increased vascular permeability enhances transendothelial transport of tissue-released chemokines. Within the vasculature, chemokines form a chemotactic gradient sequestered on heparan sulfate, which directs crawling neutrophils and expedites their extravasation to the inflamed tissue. Consequently, gradient formation grants efficient bacterial clearance. Citrullination of chemokines by leukocyte-derived PAD enzymes in the inflamed tissue prevents chemokine transport into blood vessels, which dampens further neutrophil recruitment and thereby controls the amplitude of the inflammatory response. Moreover, the mechanisms of neutrophil recruitment in response to proangiogenic factors released during hypoxia are revealed to differ from those observed during classical inflammation. Particularly, VLA-4 integrin and VEGFR1 expressed on a defined subset of neutrophils, along with endothelial VEGFR2, are required for efficient neutrophil recruitment to hypoxia. Rather than stimulus-induced phenotypic changes on neutrophils, specific neutrophil subtypes with innate proinflammatory or proangiogenic functions (respectively, CD49d-VEGFR1lowCXCR4low and CD49d+VEGFR1highCXCR4high) coexist in the circulation of humans and mice. In summary, this dissertation provides relevant information on specific steps of neutrophil recruitment to inflamed or hypoxic tissues, which may represent future means to down-regulate aberrant immune responses during chronic inflammation and autoimmune diseases; to increase angiogenesis during ischemia; or to limit pathological angiogenesis, a characteristic of tumor growth and of several chronic inflammatory disorders.

O treinamento de sistemas de Tradução de Máquina baseada em Estatística (TME) é bastante dependente da disponibilidade de corpora paralelos. Entretanto, este tipo de recurso costuma ser difícil de ser encontrado, especialmente quando lida com idiomas com poucos recursos ou com tópicos muito específicos, como, por exemplo, dermatologia. Para contornar esta situação, uma possibilidade é utilizar corpora comparáveis, que são recursos muito mais abundantes. Um modo de adquirir corpora comparáveis é a aplicação de algoritmos de Coleta Focada (CF). Neste trabalho, são propostas novas abordagens para CF, algumas baseadas em n-gramas e outras no poder expressivo das expressões multipalavra. Também são avaliadas a viabilidade do uso de CF para realização de adaptação de domínio para sistemas genéricos de TME e se há alguma correlação entre a qualidade dos algoritmos de CF e dos sistemas de TME que podem ser construídos a partir dos respectivos dados coletados. Os resultados indicam que algoritmos de CF podem ser bons meios para adquirir corpora comparáveis para realizar adaptação de domínio para TME e que há uma correlação entre a qualidade dos dois processos. / Statistical Machine Translation (SMT) is highly dependent on the availability of parallel corpora for training. However, these kinds of resource may be hard to be found, especially when dealing with under-resourced languages or very specific domains, like the dermatology. For working this situation around, one possibility is the use of comparable corpora, which are much more abundant resources. One way of acquiring comparable corpora is to apply Focused Crawling (FC) algorithms. In this work we propose novel approach for FC algorithms, some based on n-grams and other on the expressive power of multiword expressions. We also assess the viability of using FC for performing domain adaptations for generic SMT systems and whether there is a correlation between the quality of the FC algorithms and of the SMT systems that can be built with its collected data. Results indicate that the use of FCs is, indeed, a good way for acquiring comparable corpora for SMT domain adaptation and that there is a correlation between the qualities of both processes.

Le corpus est la matière première de la linguistique informatique et du traitement automatique du langage. Peu de langues disposent de corpus riches en ressources web (forums, blogs, etc.), et ce bien qu'elles soient parfois les seules disponibles. Or ces ressources contiennent beaucoup de bruit (menus, publicités, etc.). Le filtrage des données parasites et des répétitions nécessite un nettoyage à grand échelle que les chercheurs font en général à la main.Cette thèse propose un système automatique de constitution de corpus web nettoyés de leur bruit. Il est constitué de trois modules : (a) un module de construction de corpus en n'importe quelle langue et sur tout type de données, prévu pour être collaboratif et historisé ; (b) un module d'aspiration des pages web orienté sur les forums et des blogs ; (c) un module d'extraction de données pertinentes, utilisant des techniques de clustering selon différentes distances à partir de la structure de la page. Le système est évalué sous l'angle de l’efficacité de la suppression du bruit et du temps d’exécution. Nos expérimentations, faites sur quatre langues, sont évaluées à l'aide de notre propre corpus de référence. Pour mesurer la qualité, nous utilisons rappel, précision et F-mesure. Deux distances, la feature-distance et la distance de Jaro, donnent les meilleurs résultats, mais pas dans les mêmes contextes, la feature-distance ayant la meilleure qualité moyenne.Nous comparons notre méthode avec trois méthodes traitant du même problème que la nôtre, Nutch, BootCat et JusText. Les performances de notre système sont meilleures pour la qualité d’extraction, même si pour le temps de calcul, Nutch et BootCat dominent. / Corpora are the main material of computer linguistics and natural language processing. Not many languages have corpora made from web resources (forums, blogs, etc.), even those that do not have other resources. Web resources contain lots of noise (menus, ads, etc.). Filtering boilerplate and repetitive data requires a large-scale manual cleaning by the researcher.This thesis presents an automatic system that construct web corpus with a low level of noise.It consists of three modules : (a) one for building corpora in any language and any type of data, intended to be collaborative and preserving corpus history; (b) one for crawling web forums and blogs; (c) one for extracting relevant data using clustering techniques with different distances, from the structure of web page.The system is evaluated in terms of the efficacy of noise filtering and of computing time. Our experiments, made on four languages, are evaluated using our own gold standard corpus. To measure quality, we use recall, precision and F-measure. Feature-distance and Jaro distance give the best results, but not in the same contexts, feature-distance having the best average quality.We compare our method with three methods dealing with the same problem, Nutch, BootCat and JusText. The performance of our system is better as regards the extraction quality, even if for computing time, Nutch and BootCat dominate.

Title: Interactive crawling and data extraction Author: Bc. Petr Fejfar Author's e-mail address: pfejfar@gmail.com Department: Department of Distributed and Dependable Systems Supervisor: Mgr. Pavel Je ek, Ph.D., Department of Distributed and De- pendable Systems Abstract: The subject of this thesis is Web crawling and data extraction from Rich Internet Applications (RIA). The thesis starts with analysis of modern Web pages along with techniques used for crawling and data extraction. Based on this analysis, we designed a tool which crawls RIAs according to the instructions defined by the user via graphic interface. In contrast with other currently popular tools for RIAs, our solution is targeted at users with no programming experience, including business and analyst users. The designed solution itself is implemented in form of RIA, using the Web- Driver protocol to automate multiple browsers according to user-defined instructions. Our tool allows the user to inspect browser sessions by dis- playing pages that are being crawled simultaneously. This feature enables the user to troubleshoot the crawlers. The outcome of this thesis is a fully design and implemented tool enabling business user to extract data from the RIAs. This opens new opportunities for this type of user to collect data from Web pages for use...

Muitas informações disponíveis na Web estão armazenadas em bancos de dados on-line e são acessíveis somente após um usuário enviar uma consulta por meio de uma interface de busca. Essas informações estão localizadas em uma parte da Web conhecida como Web Oculta ou Web Profunda e, geralmente, são inacessíveis por máquinas de busca tradicionais. Uma vez que a forma de acessar os dados na Web Oculta se dá por intermédio de submissões de consultas, muitos trabalhos têm focado em como preencher automaticamente campos de formulários. Esta tese apresenta uma metodologia para o preenchimento de formulários na Web Oculta. Além disso, descreve uma categorização das técnicas de preenchimento de formulários existentes no estado da arte de coleta na Web Oculta, produzindo uma análise comparativa entre elas. A solução proposta descreve um método automático para seleção de valores para campos de formulários combinando heurísticas e técnicas de aprendizagem de máquina. Experimentos foram realizados em formulários reais da Web, de vários domínios, e os resultados indicam que a abordagem proposta apresenta desempenho comparável aos obtidos pelas técnicas do estado da arte, sendo inclusive significativamente diferente com base em avaliação estatística. / A large portion of the information on the Web is stored inside online databases. Such information is accessible only after the users submit a query through a search interface. TheWeb portion in which that information is located is called HiddenWeb or DeepWeb, and generally this part is inaccessible by traditional search engines crawlers. Since the only way to access the Hidden Web pages is through the query submissions, many works have focused on how to fill in form fields automatically, aiming at enhancing the amount of distinct information hidden behind Web forms. This thesis presents an automatic solution to value selection for fields in Web forms. The solution combines heuristics and machine learning techniques for improving the selection of values. Furthermore, this proposal also describes a categorization of form filling techniques and a comparative analysis between works in the state of the art. Experiments were conducted on real Web sites and the results indicated that our approach significantly outperforms a baseline method in terms of coverage without additional computational cost.

Submitted by Pós-Com Pós-Com (pos-com@ufba.br) on 2015-04-17T15:06:44Z No. of bitstreams: 1 Talyta Louise Todescat Singer - Dissertação.pdf: 4034000 bytes, checksum: dffc645dc88549ae313a8951db3d574f (MD5) / Approved for entry into archive by Vania Magalhaes (magal@ufba.br) on 2017-09-29T16:25:19Z (GMT) No. of bitstreams: 1 Talyta Louise Todescat Singer - Dissertação.pdf: 4034000 bytes, checksum: dffc645dc88549ae313a8951db3d574f (MD5) / Made available in DSpace on 2017-09-29T16:25:19Z (GMT). No. of bitstreams: 1 Talyta Louise Todescat Singer - Dissertação.pdf: 4034000 bytes, checksum: dffc645dc88549ae313a8951db3d574f (MD5) / CNPQ / Esta pesquisa se dedica a realizar um estudo exploratório da internet das coisas, identificando e descrevendo os tensionamentos surgidos da produção, captura, processamento e/ou transmissão de informação por objetos interconectados observáveis a partir dos rastros digitais públicos. Nos concentramos em identificar questões sensíveis, temas que levantem discussões e mobilizem diferentes tipos de atores – desenvolvedores, políticos, usuários, indústria, leis, protocolos – e possam ser compreendidas por mais de um ponto de vista. A pesquisa tem como marco teórico a Teoria Ator-Rede que a partir de seu princípio de simetria entre humanos e não-humanos que confere a ambos a possibilidade de agência. A metodologia empregada é a de cartografia de controvérsias, um conjunto de técnicas aplicáveis a exploração e visualização de conflitos. A pesquisa empírica é formada por um mapeamento da rede temática da internet das coisas a partir de web crawling de sites em português e em inglês e análise de conteúdo de notícias publicadas sobre o assunto em sites de notícia de grande visibilidade. A pesquisa identificou controvérsias em seis temas: dependência tecnológica, software livre, padronização, legislação, privacidade e segurança. A dissertação está dividida em três capítulos: o primeiro dedicado a criar um panorama atual da internet das coisas, discutir conceitos e apresentar uma linha do tempo; o segundo apresenta conceitos-chave da Teoria Ator-Rede e as etapas da cartografia de controvérsias; o último capítulo explicita as etapas de coleta e processamento de dados e os principais resultados. Nas considerações finais apresentamos um quadro síntese das controvérsias encontradas e uma árvore de argumentos que identifica os principais pontos de discordância e os principais atores que participam das discussões em torno dos objetos conectados. / The present research is an exploratory study of internet of things that identifies and describes the conflicts emerging on connected objects production, processing and transmission of information. We are concerned about the sensible questions that involve different kinds of actants -users, developers, politicians, industry, laws, communication protocols - and needs to be observed by multiple viewpoints. Actor - Network Theory is our theoretical framework that includes the principle of generalized symmetry be tween humans and non-humans and gives them both agency capacity. We use the cartography of controversies as a method to explore and visualize public debates through a set of techniques that includes web crawling and content analysis. Our results show controversies about six subjects: internet addiction, free software, standardization, legislation, privacy and security. The research is divides in three chapters: the first one creates a landscape of internet of things, its definition and history; the second one presents the key concepts about Actor-Network Theory and the layers of cartography of controversies; the last chapter reports our methodological choices and main results. Our final remarks show a summary table of controversies found and a disagreement tree.