Age related macular degeneration (AMD) is a public health concern in an aging society. The retinal pigment epithelium (RPE) layer of the eye is a principal site of pathogenesis for AMD. Morphological characteristics of the cells in the RPE layer can be used to discriminate age and disease status of individuals. In this thesis three genotypes of mice of various ages are used to study the predictive abilities of these characteristics. The disease state is represented by two mutant genotypes and the healthy state by the wild-type. Classification analysis is applied to the RPE morphology from the different spatial regions of the RPE layer. Variable reduction is accomplished by principal component analysis (PCA) and classification analysis by the k-nearest neighbor (k-NN) algorithm. In this way the differential ability of the spatial regions to predict age and disease status by cellular variables is explored.

Nutriomics is the study of the whole range of nutritional components (nutriome) in foods. In order to further understand the molecular basis for the positive health benefits of fruits identified from epidemiology, the mass balance through the human digestion and absorption should be studied. The components of the nutriome studied in this research were sugars, carotenoids, phenolics and organic acids, all important for defining dietary – human health relationships and linking to evidence obtained from epidemiological studies. An attempt to approach a realistic human mimetic digestion and absorption model has been carried out in this study using a static in-vitro model of the human digestive system. Two major novelities in this model compared to other in-vitro models are (i) the use of particles of solid fruit products that mimic the products of human chewing and (ii) a cell-based (Caco-2) in-vitro intestinal absorption model. Hence, imitative bioavailability, i.e. releasing nutrients and potential levels of target compounds reaching the portal circulatory system could be assessed. The fruits studied were tomato, mango, papaya; each as fresh, dried and juiced forms. In-vivo chewing suggested 0.5 cm size modes for dried products and 1.5 cm for fresh products. The agglomerates that were obtained from the chewing of dried products disaggregated during in-vitro digestions in the presence of acids (gastric simulation) or sodium bicarbonate at pH 6 (small intestinal simulation). The extent of this disaggregation followed the order: tomato > mango > papaya. Although all fresh samples contained separated cells, their responses to a 5 mm texture analysis probe (mimicking teeth cusps) varied depending on fruit products. All matrices were hardened by drying, becoming more brittle and breaking easier to produce smaller size modes. Variation between individual participants in the size of their chewed particles was lower for fresh products and high for dried products. The in-vitro digestion and absorption model developed had simulated particle sizes of approximately 0.5 cm3 for dried products or 1.5 cm3 (thickness varied with the products) for fresh products in a 9:1 ratio mix with blended samples, and were digested in-vitro using the following steps: 1. ‘Chewing’: pH 6.9; 37 C, 10 min, in a shaking-water bath (55 rpm) with human alpha-amylase (100 U/L). 2. ‘Gastric’ digestion: pH 2; 37 C, 60 min, in a shaking-water (55 rpm) with porcine pepsin (40 µg/L). 3. ‘Intestinal’ digestion: pH 6; 37 C, 60 min, in a shaking-water bath (55 rpm) with porcine pancreatin and bile extract (1.4 µg/L and 8.6 µg/L, respectively). 4. Caco-2 cell monolayer in-vitro passages: aged 22 days post confluent monolayers in a 24 transwell-insert well plate seeded at 105 cells, pH 7.4 with renewal of apical and basolateral solutions every 30 min for bioavailability estimations. In this study, two models of basolateral – apical solution renewals were carried out: both apical and basolateral were renewed (model A) and basolateral only was renewed (model B). To study metabolites produced by Caco-2 cells, the bioassays were carried out for 22 h without renewals of apical and basolateral solutions (model C). An overview of nutriomics analysis of in-vitro digestions of mango, papaya and tomato based on principal component analysis (PCA) suggested: (1) fruit types led to variable nutriome releases: in-vitro digestions affected tomato >mango >papaya; (2) processing varied nutriome releases from fruit products with juicing tended to release more nutriome components, whereas drying and unprocessed (fresh) did not show noticeably different patterns; (3) gastric and simultaneous gastric-intestinal digestions were similar in nutriome releases whereas contributions of intestinal digestion alone were negligible for water soluble nutriome components; and overall (4) during in-vitro digestions there were no interactions among releasing nutriome from the fruit products studied (independent nutriome releasing processes). Phenolic components showed molecular changes during in-vitro digestion and processing, due to, heating effects, pH or enzymic degradations. Caco-2 bioassays using model compounds showed a range of monolayer responses as follows: (1) mannitol, lycopene and catechin were strictly retained in the apical solution; (2) sugars, caffeine and atenolol were translocated in the apical-to-basolateral direction as intact molecules; (3) Beta-carotene partially disappeared from the apical solution without basolateral release. Models A – C consistently confirmed these responses. Low recoveries provided evidence for cellular metabolisms of (particularly) phenolic and carotenoid molecules by the Caco-2 cell monolayers.

Made available in DSpace on 2016-06-02T19:32:09Z (GMT). No. of bitstreams: 1 5793.pdf: 2202297 bytes, checksum: 747359b5bd140047bbb32daf90dd91aa (MD5) Previous issue date: 2014-02-27 / Financiadora de Estudos e Projetos / Environmental contamination by pesticides in agriculture has caused many irreparable damage environmental to the ecosystem. Pentachlorophenol (PCP), the class of organochlorines, is a phenolic compound and a dangerous pollutant, that although banned in Brazil since 1985, there are many contaminated areas. This pesticide is the subject of much concern because it has high toxicity and power of persistence in the environment due to their resistance to biotic and abiotic degradation. The use of microorganisms as degrading agents of several chemical is considered an effective method to reduce the adverse effects of contaminants on the environment. The fungi derived from marine environment are adapted to extreme conditions, developing attributes that give them the ability to produce various of biologically active compounds differentes of their respective representatives who inhabit the land environment. In addition, marine fungi present an excellent bioenzimatic potential to be explored in the biotransformation of xenobiotics such as, for example, pesticides. In this work, fifteen strains of fungi isolated from a marine invertebrate, the ascidian Didemnun ligulum were evaluated according to their resistance, ability to grow in the presence of the pentachlorophenol (PCP) pesticide and its enzymatic potential against biodegradation of the pesticide and its metabolites. Concentrations were evaluated 10, 25, 30, 40 and 50 mgL-1 in solid media 3% malt. Nine among the tested strains showed growth in at least one concentration, but the DL2B strain (identified as Trichoderma harzianum) obtained optimal growth in most (50 mgL-1), proving be resistant to its toxicity and suggesting its potential for biodegradation. Therefore, it was selected for reactions in liquid medium in the presence of the pesticide with an initial concentration of 20 mgL-1 of PCP, to measure biodegradation. he biodegradation were evaluated after 7, 14 and 21 days of incubation. In 7 days of incubation it was no longer detected the presence of PCP in the samples, indicating the biodegradation of the pesticide by the fungus. The pentachloroanisole (PCA) metabolites and 2,3,4,6- tetrachloroanisole (2,3,4,6-TeCA) were identified by gas chromatography coupled with mass spectrometry (GC-MS), indicating degradation of the pesticide by the fungus. In a second step, the strain of Trichoderma harzianum was subjected to biodegradation reactions in liquid medium containing metabolites PCA and 2,3,4,6-TeCA. It was observed that these compounds possibly cause some toxic effect on this micro-organism, but not enough to cripple the biodegradation by fungi. The fungus didn't degrade these compounds completely as occured with the PCP, but got reduce the concentrations of metabolites in the samples. These results confirm the efficiency of fungi derived from marine environment to biodegrade persistent compounds and enables the improvement methodologies using these microorganisms in techniques for decontamination of polluted environments with xenobiotics. / A contaminação ambiental pelo uso de pesticidas na agricultura tem causado muitos danos ambientais irreparáveis ao ecossistema. O pentaclorofenol (PCP), da classe dos organoclorados, é um composto fenólico e um perigoso poluente. Este pesticida é objeto de muita preocupação por possuir elevada toxicidade e poder de persistência no meio ambiente, devido à sua resistência à degradação biótica e abiótica. Apesar de proibido no Brasil desde 1985, existem ainda muitas áreas contaminadas por ele. A utilização de micro-organismos como agentes degradadores de diversas substâncias químicas é considerada uma forma eficiente para reduzir os efeitos adversos dos contaminantes sobre o ambiente. Os fungos derivados do ambiente marinho são adaptados à condições extremas e possuem atributos que lhes conferem a capacidade de produzir compostos biologicamente ativos diferentes dos seus respectivos representantes que habitam o meio terrestre. Além disso, os fungos marinhos apresentam um excelente potencial bioenzimático a ser explorado na biotransformação de xenobióticos como, por exemplo, os pesticidas. Neste trabalho, quinze linhagens de fungos isoladas de um invertebrado marinho, a ascídia Didemnun ligulum, foram avaliadas segundo sua resistência à toxicidade, habilidade de crescer e biodegradar o pesticida pentaclorofenol (PCP) e seus metabólitos. Foram avaliadas as concentrações 10, 25, 30, 40 e 50 mgL-1 de PCP, em meio de cultura sólido Malte 3%. Nove, dentre as linhagens avaliadas, apresentaram crescimento em pelo menos uma das concentrações, mas a linhagem DL2B (identificada como Trichoderma harzianum) apresentou ótimo crescimento na concentração mais drástica (50 mgL-1), mostrando ser resistente à sua toxicidade e sugerindo seu potencial para biodegradação do composto. Sendo assim, foi selecionada para reações em meio líquido na presença do pesticida com concentração inicial de 20 mgL-1 de PCP, para quantificar sua biodegradação. As avaliações da biodegradação foram realizadas após 7, 14 e 21 dias de incubação. Com 7 dias de incubação já não foi detectada a presença do PCP nas amostras. Os metabólitos pentacloroanisol (PCA) e 2,3,4,6-tetracloroanisol (2,3,4,6-TeCA) oriundos das biotransformação do PCP foram identificados por cromatografia a gás acoplada ao espectrômetro de massas (CG-EM), indicando a biotransformação do pesticida pelo fungo. Em uma segunda etapa, a linhagem de Trichoderma harzianum foi submetida à reações de biodegradação em meio líquido contendo os metabólitos PCA e 2,3,4,6-TeCA. Observou-se que esses compostos possivelmente causam algum efeito tóxico sobre este micro-organismo, mas não o suficiente para inviabilizar a biodegradação pelo fungo. O fungo não degradou esses compostos completamente no período de 21 dias, como ocorreu com o PCP, mas conseguiu reduzir as concentrações dos metabólitos nas amostras. Estes resultados confirmam a eficiência de fungos derivados do ambiente marinho em biodegradar compostos persistentes e possibilitam o aprimoramento de metodologias utilizando esses micro-organismos em técnicas para descontaminação de ambientes poluídos com xenobióticos.

Les avancées des nouvelles techniques de séquençage ont permis de produire des données hétérogènes, volumineuse, de grande dimension et à différentes échelles du vivant. L'intégration de ces différentes données représente un défi en biologie des systèmes, défi qu'il est critique d'aborder pour tirer le meilleur parti possible de l'accumulation d'informations biologiques pour leur interprétation et leur exploitation dans un but finalisé. Cette thèse regroupe plusieurs contributions méthodologiques utiles à l'exploration simultanée de plusieurs jeux de données omiques de natures hétérogènes. Pour aborder cette question, les noyaux et les méthodes à noyaux offrent un cadre naturel, car ils permettent de prendre en compte la nature propre de chacun des tableaux de données tout en permettant leur combinaison. Toutefois, lorsque le nombre d'observations à traiter est grand, les méthodes à noyaux souffrent d'un manque d'interprétabilité et d'une grande complexité algorithmique. Une première partie de mon travail a porté sur l'adaptation de deux méthodes exploratoires à noyaux : l'analyse en composantes principales (K-PCA) et les cartes auto- organisatrices (K-SOM). Les adaptations développées portent d'une part sur le passage à l'échelle du K-SOM et de la K-PCA au domaine des omiques et d'autre part sur l'amélioration de l'interprétabilité des résultats. Dans une seconde partie, je me suis intéressé à l'apprentissage multi-noyaux pour combiner plusieurs jeux de données omiques. L'efficacité des méthodes proposées est illustrée dans le contexte de l'écologie microbienne : huit jeux de données du projet TARA oceans ont été intégrés et analysés à l'aide d'une K-PCA. / The development of high-throughput sequencing technologies has lead to produce high dimensional heterogeneous datasets at different living scales. To process such data, integrative methods have been shown to be relevant, but still remain challenging. This thesis gathers methodological contributions useful to simultaneously explore heterogeneous multi-omics datasets. To tackle this problem, kernels and kernel methods represent a natural framework because they allow to handle the own nature of each datasets while permitting their combination. However, when the number of sample to process is high, kernel methods suffer from several drawbacks: their complexity is increased and the interpretability of the model is lost. A first part of my work is focused on the adaptation of two exploratory kernel methods: the principal component analysis (K-PCA) and the self-organizing map (K-SOM). The proposed adaptations first address the scaling problem of both K-SOM and K-PCA to omics datasets and second improve the interpretability of the models. In a second part, I was interested in multiple kernel learning to combine multiple omics datasets. The proposed methods efficiency is highlighted in the domain of microbial ecology: eight TARA oceans datasets are integrated and analysed using a K-PCA.

Dans cette thèse, nous nous intéressons au problème de la segmentation de portraits numériques. Nous appelons portrait numérique la photographie d’une personne avec un cadre allant grossièrement du gros plan au plan poitrine. Le problème abordé dans ce travail est un cas spécifique de la segmentation d’images où il s’agit notamment de définir précisément la frontière de la région « cheveux ». Ce problème est par essence très délicat car les attributs de la région « cheveux » (géométrie, couleur, texture) présentent une grande variabilité à la fois entre les personnes et au sein de la région. Notre cadre applicatif est un système d’« essayage virtuel » de lunettes à destination du grand public, il n’est pas possible de contrôler les conditions de prise de vue comme l’éclairage de la scène ou la résolution des images, ce qui accroît encore la diculté du problème. L’approche proposée pour la segmentation de portraits numériques est une approche du plus grossier au plus fin procédant par étapes successives. Nous formulons le problème comme celui d’une segmentation multi-régions, en introduisant comme « régions secondaires », les régions adjacentes à la région « cheveux » , c.-à-d. les régions « peau » et « fond ». La méthode est fondée sur l’apparence (appearance-based method) et a comme spécificité le fait de déterminer les descripteurs de régions les plus adaptés à partir d’une base d’images d’apprentissage et d’outils statistiques. À la première étape de la méthode, nous utilisons l’information contextuelle d’un portrait numérique — connaissances a priori sur les relations spatiales entre régions— pour obtenir des échantillons des régions « cheveux », « peau » et « fond ». L’intérêt d’une approche fondée sur l’apparence est de pouvoir s’adapter à la fois aux conditions de prises de vue ainsi qu’aux attributs de chaque régions. Au cours de cette étape, nous privilégions les modèles de forme polygonaux couplés aux contours actifs pour assurer la robustesse du modèle. Lors de la seconde étape, à partir des échantillons détectés à l’étape précédente, nous introduisons un descripteur prenant en compte l’information de couleur et de texture. Nous proposons une segmentation grossière par classification en nous appuyant à nouveau sur l’information contextuelle : locale d’une part grâce aux champs de Markov, globale d’autre part grâce à un modèle a priori de segmentation obtenu par apprentissage qui permet de rendre les résultats plus robustes. La troisième étape ane les résultats en définissant la frontière des « cheveux » comme une région de transition. Cette dernière contient les pixels dont l’apparence provient du mélange de contributions de deux régions (« cheveux »et « peau » ou «fond »). Ces deux régions de transition sont post-traitées par un algorithme de «démélange » (digital matting) pour estimer les coecients de transparence entre « cheveux » et « peau », et entre « cheveux » et « fond ». À l’issue de ces trois étapes, nous obtenons une segmentation précise d’un portrait numérique en trois « calques », contenant en chaque pixel l’information de transparence entre les régions « cheveux », « peau » et « fond ». Les résultats obtenus sur une base d’images de portraits numériques ont mis en évidence les bonnes performances de notre méthode. / In this thesis, we are interested in the problem of the segmentation of digital portraits. We call digital portrait the photography of a person with a frame roughly ranging from the close-up to the chest plane. The problem addressed in this work is a specific case of the segmentation of images where it is especially necessary to define precisely the border of the "hair" region. This problem is inherently very delicate because the attributes of the "hair" region (geometry, color, texture) present an important variability between people and within the region. Our application is a system of "virtual fitting" of glasses for the general audience, it is not possible to control the shooting conditions such as stage lighting or image resolution, which increases the difficulty of the problem. The approach proposed for the segmentation of digital portraits is an approach « coarse to fine », proceeding in successive stages. We formulate the problem as a multi-region segmentation, introducing as "secondary regions" regions adjacent to the "hair" region, ie, the "skin" and "background" regions. The method is based on appearance-based method and has as a specificity the determination of the descriptors of regions most adapted from a database of learning and statistical tools. In the first step of the method, we use the contextual information of a Digital portrait - a priori knowledge about the spatial relations between regions - to obtain samples of the regions "hair", "skin" and "background". The value of an appearance-based approach is to be able to adapt to both the shooting conditions and the attributes of each region. During this stage, we prefer polygonal shape models coupled with active contours to ensure the robustness of the model. In the second step, from the samples detected in the previous step, we introduce a descriptor taking into account the color and texture information. We propose a rough segmentation by classification by relying on the contextual information: local on the one hand thanks to the Markov fields, global on the other hand thanks to an a priori model of segmentation obtained by learning which il allow to obtain robust results. The third stage refines the results by defining the border of "hair" as a transition region. This région contains the pixels whose appearance comes from the mixture of contributions of two regions ("hair" and "skin" or "background"). These two transition regions are post-processed by a digital matting algorithm to estimate the coefficients of transparency between "hair" and "skin", and between "hair" and "background". At the end of these three steps, we obtain a precise segmentation of a digital portrait into three "layers", containing in each pixel the information of transparency between the regions "hair", "skin" and "background". The results obtained on the basis of images of digital portraits have highlighted the good performance of our method.

Submitted by Rodrigo Martins Cruz (rodrigo.cruz@ufvjm.edu.br) on 2015-02-19T13:03:29Z No. of bitstreams: 5 nascilaine.pdf: 1827154 bytes, checksum: b9e52778b31d948210ff6f1f431250cf (MD5) license_url: 52 bytes, checksum: 3d480ae6c91e310daba2020f8787d6f9 (MD5) license_text: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) license_rdf: 23898 bytes, checksum: e363e809996cf46ada20da1accfcd9c7 (MD5) license.txt: 2110 bytes, checksum: b4c884761e4c6c296ab2179d378436d4 (MD5) / Approved for entry into archive by Rodrigo Martins Cruz (rodrigo.cruz@ufvjm.edu.br) on 2015-02-20T09:45:45Z (GMT) No. of bitstreams: 5 nascilaine.pdf: 1827154 bytes, checksum: b9e52778b31d948210ff6f1f431250cf (MD5) license_url: 52 bytes, checksum: 3d480ae6c91e310daba2020f8787d6f9 (MD5) license_text: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) license_rdf: 23898 bytes, checksum: e363e809996cf46ada20da1accfcd9c7 (MD5) license.txt: 2110 bytes, checksum: b4c884761e4c6c296ab2179d378436d4 (MD5) / Made available in DSpace on 2015-02-20T09:45:45Z (GMT). No. of bitstreams: 5 nascilaine.pdf: 1827154 bytes, checksum: b9e52778b31d948210ff6f1f431250cf (MD5) license_url: 52 bytes, checksum: 3d480ae6c91e310daba2020f8787d6f9 (MD5) license_text: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) license_rdf: 23898 bytes, checksum: e363e809996cf46ada20da1accfcd9c7 (MD5) license.txt: 2110 bytes, checksum: b4c884761e4c6c296ab2179d378436d4 (MD5) Previous issue date: 2013-02-21 / Na primeira parte do trabalho foi proposto o reconhecimento de padr?es de jabuticaba (Myrciaria jabuticaba (vell) berg (jabuticaba sabar?)) pela avalia??o dos elementos boro, nitrog?nio, magn?sio, f?sforo, pot?ssio, c?lcio, mangan?s, ferro, cobre e zinco na casca empregando-se as t?cnicas de titula??o, espectroscopia de absor??o at?mica e molecular. Os dados obtidos pelas an?lises dos elementos foram dispostos em matriz 7 x 10, sendo sete amostras referente ?s sete cidades analisadas (S?o Jo?o da Chapada, Gouveia, Guinda, Couto de Magalh?es, Po?os de Caldas, Belo Horizonte e Serro) e dez vari?veis correspondendo ?s concentra??es dos elementos determinadas em mgk-1. O reconhecimento de padr?es foi feito empregando-se a An?lise das Componentes Principais e An?lise Hier?rquica por Agrupamentos. Desta forma foi poss?vel identificar pela An?lise das Componentes Principais os grupos formados pela semelhan?a, baseados nos pesos correspondentes ?s influ?ncias das vari?veis. Cerca de cinco componentes principais acumularam 97,47% da vari?ncia total dos dados sendo que a componente principal de n?mero um (01) acumulou cerca de 38,21% da vari?ncia e a componente principal de n?mero tr?s (03) cerca de 16,13% da vari?ncia. Os grupos formados na An?lise das Componentes Principais foram confirmados na An?lise Hier?rquica por Agrupamento, sendo formado quatro grupos, denotando a maior semelhan?a nas amostras de Gouveia e Couto Magalh?es. Na segunda etapa do estudo teve por finalidade classificar e caracterizar quimicamente o refrigerante do tipo guaran?, em diferentes estados brasileiros, foram analisadas suas concentra??es de metais e diferentes compostos atrav?s da absor??o at?mica e espectroscopia de infravermelho respectivamente. Para classificar os refrigerantes dos diversos locais, foi utilizada nos resultados obtidos das an?lises a an?lise explorat?ria baseada em An?lise das Componentes Principais (PCA). A melhor separa??o se deu com as vari?ncias acumuladas na CP2 e CP4, onde as amostras de refrigerantes zero formaram um grupo e amostras coletadas no RJ outro grupo e as amostras de Shim tamb?m formaram um grupo diferente dos demais. Como conclus?o pode-se salientar que os m?todos de reconhecimento de padr?es empregados neste trabalho apresentaram resultados que possibilitam a diferencia??o das amostras em locais distintos. / Disserta??o (Mestrado) ? Programa de P?s-Gradua??o em Qu?mica, Universidade Federal dos Vales do Jequitinhonha e Mucuri, 2012. / ABSTRACT In the first part of the proposed pattern recognition Jabuticaba (Myrciaria jabuticaba (vell) berg (jabuticaba sabar?)) for the evaluation of the elements boron, nitrogen, magnesium, phosphorus, potassium, calcium, manganese, iron, copper and zinc on the shell employing the techniques of titration, atomic absorption spectroscopy and molecular. The data obtained by analysis of the elements were arranged in a matrix 10 x 7, seven samples analyzed referring to the seven cities (S?o Jo?o da Chapada, Gouveia, Guinda, Couto de Magalhaes, Wells Caldas, Belo Horizonte and Serro) and ten variables corresponding concentrations of certain elements in MGK-1. Pattern recognition was done using the Analysis of Principal Components Analysis for Hierarchical Groups. This made it possible to identify the Principal Component Analysis groups formed by similarity, based on the weights corresponding to the influences of variables. About five main components accumulated 97.47% of the total variance of the data being the main component of a number (01) accumulated approximately 38.21% of the variance and the principal component number three (03) approximately 16.13% variance. The groups formed in the Principal Component Analysis were confirmed in the Hierarchic Cluster Analysis, formed four groups, indicating a greater similarity in the samples of Gouveia and Couto Magalh?es. In the second stage of the study aimed to classify and characterize chemically the guarana soda type, in different Brazilian states were analyzed in relation to different concentrations of metals and compounds by atomic absorption and infrared spectroscopy respectively. To sort the drinks at different locations, we used the results of the analyzes exploratory analysis based on Principal Component Analysis (PCA). The best separation was accumulated at variance with the CP2 and CP4, where samples of soft drinks have formed a group and zero samples collected at the RJ and the other group samples Shim also formed a different group of demais. Como conclusion can be noted that methods of pattern recognition in this work presented results that allow the differentiation of samples at different locations.

Ao se discutir a calçada como o “cenário”, onde acontecem as cenas urbanas de uma cidade, e seus “atores” – a população usuária – é possível entender os conflitos gerados pela escolha de um determinado piso, localização de um mobiliário, possibilidade de uma aglomeração, existência de uma vegetação, delimitação diferenciada de seus usuários. Ao se avaliarem as calçadas em conjunto com seu paisagismo, seu mobiliário, sua distribuição de uso e seus fatores históricos, é possível identificar conceitos anteriores de pensamento (e necessidade) urbano que não atendem aos desejos de hoje. As situações acima descritas aparecerão com a falta de controle no planejamento dos espaços urbanos e na permissividade desenfreada de ações particulares sobre o uso coletivo. São conflitos que gerarão problemas relativos à segurança dos usuários, ao impedimento de uso dos espaços públicos, à simples circulação das pessoas. O estudo e interpretação de determinados conjuntos de dados envolve muitas vezes a análise de um grande número de variáveis, constituindo-se em um trabalho difícil e demorado. O emprego de métodos multivariados, como a Análise de Componentes Principais (ACP), destaca-se como uma ferramenta bastante útil quando o número de variáveis é grande. A ACP destina-se à análise de dados quantitativos e este trabalho se propôs a descrever e avaliar o uso de uma análise quantitativa em conjunto com dados qualitativos. As técnicas são aplicadas a um estudo de caso dirigido na área central da cidade de Foz do Iguaçu com o objetivo de avaliar o nível de serviço das calçadas existentes através de uma análise conjunta de fatores qualitativos e quantitativos, visando a melhoria da acessibilidade dos pedestres em geral. / In discussing the sidewalk as the “scenario”, where the scenes take place in an urban city, and its “players” – the user population – it is possible to understand the conflicts generated by choosing a particular floor, a location of furniture, possibility of conurbation, the existence of vegetation, differentiated demarcation of its users. When evaluating the sidewalks in conjunction with your landscaping, your furniture, your use and distribution of their historical factors can be identified earlier concepts of thought (and need) that do not meet city desires today. The situations described above appear with the lack of control in the planning of urban spaces and unbridled permissiveness of particular actions on collective use. These are conflicts that generate safety issues for users, unable to use public spaces, the simple movement of persons. The interpretation and study of certain sets of data often involves the analysis of a large number of variables, being in a difficult and time consuming. The use of multivariate methods such as Principal Component Analysis (PCA), stands out as very useful tool when the number of variables is large. The PCA is intended for the analysis of quantitative data and this work was to describe and evaluate the use of a quantitative analysis in conjunction with qualitative data. The techniques are applied to a case study aimed at the central area of the city of Foz do Iguaçu in order to assess the level of service of existing sidewalks through a joint analysis of qualitative and quantitative factors in order to improve the accessibility of pedestrian general.

Submitted by Ana Hilda Fonseca (anahilda@ufba.br) on 2013-04-09T13:35:48Z No. of bitstreams: 1 dissertação.15.12[1].pdf: 2814994 bytes, checksum: 036ac54999fb7e9e04a62a51d84ccaed (MD5) / Approved for entry into archive by Ana Hilda Fonseca(anahilda@ufba.br) on 2013-04-19T12:41:43Z (GMT) No. of bitstreams: 1 dissertação.15.12[1].pdf: 2814994 bytes, checksum: 036ac54999fb7e9e04a62a51d84ccaed (MD5) / Made available in DSpace on 2013-04-19T12:41:43Z (GMT). No. of bitstreams: 1 dissertação.15.12[1].pdf: 2814994 bytes, checksum: 036ac54999fb7e9e04a62a51d84ccaed (MD5) Previous issue date: 2010 / CAPES / Fontes de poluição podem comprometer os diversos fins de um manancial, principalmente o fim mais nobre, o consumo humano. Este trabalho tem como objetivo, avaliar e interpretar os fatores físicos, químicos e biológicos que interferem nas características de um corpo hídrico,nesse caso específico da Bacia Hidrográfica do Rio Paraguaçu, na Bahia. Para realizar este trabalho foi aplicada análise exploratória multivariada, Análise de Componentes Principais - PCA e Análise de Agrupamento Hierárquica - HCA, nos dados disponibilizados pelo Instituto de Gestão das Águas e Clima - INGÁ, correspondendo às amostras coletadas em 55 pontos e analisadas quanto aos parâmetros fisico-quimicos: temperatura, pH, turbidez, sólidos totais, oxigênio dissolvido (OD), sulfato, carbono orgânico total (COT), demanda bioquímica de oxigênio (DBO), Demanda Química de Oxigênio (DQO), nitrato, amônia, nitrogênio total, fósforo total, surfactantes, ferro, manganês, bário, níquel e zinco; e microbiológico: coliformes termotolerantes. Utilizando-se a PCA e a HCA como ferramentas estatisticas para avaliar os dados obtidos dos parâmetros físico-químicos e microbiológico, determinou-se os pontos de amostragem que apresentaram um comportamento atípico (outleirs), através do agrupamento dos pontos em função dos escores, além das variáveis que justificaram esse comportamento por meio dos pesos. A PCA promoveu uma redução de 20 dimensões (variáveis) para 3 componentes principais de acordo com os valores da variância de cada, tornando o sistema mais simples de ser analisado. Com a HCA, constatou-se agrupamentos de alguns pontos de amostragem, ratificando alguns agrupamentos obtidos com a PCA. Sendo assim, neste trabalho, a análise multivariada de dados facilitou bastante a determinação dos pontos de amostragem atípicos (outleirs), assim como as variáveis que justificaram esse comportamento. Com isso, pôde-se ter um conhecimento mais apurado das condições ambientais da Bacia Hidrográfica do Rio Paraguaçu, podendo favorecer a tomada de medidas corretivas e preventivas para que se tenha água em quantidade e qualidade satisfatórias para consumo humano. / Salvador

Submitted by Paula Andrea Soto (paulaandreasoto@hotmail.com) on 2016-09-05T12:30:23Z No. of bitstreams: 1 Paula Andrea Soto Dissertacao.pdf: 4060630 bytes, checksum: a38f57b1ee13eb3c036f96d824c204fe (MD5) / Approved for entry into archive by Renata de Souza Nascimento (renata.souza@fgv.br) on 2016-09-05T18:26:52Z (GMT) No. of bitstreams: 1 Paula Andrea Soto Dissertacao.pdf: 4060630 bytes, checksum: a38f57b1ee13eb3c036f96d824c204fe (MD5) / Made available in DSpace on 2016-09-05T18:28:40Z (GMT). No. of bitstreams: 1 Paula Andrea Soto Dissertacao.pdf: 4060630 bytes, checksum: a38f57b1ee13eb3c036f96d824c204fe (MD5) Previous issue date: 2016-08-11 / Ao modelar séries de preços de ativos financeiros, a prática usual é tomar a primeira diferença das séries, e trabalhar assim com retornos ou logretornos. Utilizando VECM (Vector Error Correction Models, em inglês), torna-se possível trabalhar diretamente com as séries sem diferenciar, o que possibilita o estudo de tendências comuns e cointegração. Este trabalho utiliza VECM para gerar estratégias de arbitragem estatística no mercado brasileiro de ações. Tendências comuns são identificadas por PCA (Principal Components Analysis, em inglês, ou análise de componentes principais, em português) e os resultados foram utilizados para definir portfólios cointegrados. Foram propostos dois métodos de geração de sinais para estratégias de trading do tipo longshort. Um total de cinco diferentes estratégias de trading foram simuladas e a existência de arbitragem estatística em cada caso foi testada pelo teste proposto em (JARROW et al., 2012). Conclui-se que, ao considerar séries de preços não diferenciadas, a metodologia abordada permite identificar e modelar candidatos de portfólios cointegrados. Quando bem calibradas, as estratégias testadas geram ganhos significativos em todos os portfólios. / Common practice for modelling stock prices is to use their differences in form of returns or logreturns. Using VECM (Vector Error Correction Models), it is possible to work with the series of prices without differentiation, which allows looking into common trends and cointegration. This work uses VECM to create trading strategies for the Brazilian stock market. Common trends are obtained using PCA (Principal Components Analysis) and prices are modelled using VECM. Five longshort-type trading strategies are simulated in diversified portfolios, and tested for statistical arbitrage using the test proposed by (JARROW et al., 2012). The methodology for identifying common trends and modelling prices allows for trading strategies with good results for all portfolios.

As inferências têm diversas aplicações na indústria de processos químicos, sendo essenciais no sucesso de projetos de controle avançado. O desempenho do controle será sempre ligado ao desempenho da inferência, sendo importante a manutenção da sua qualidade ao longo do tempo. Neste trabalho, uma nova metodologia é sugerida para o desenvolvimento de inferências baseadas em dados seguindo uma abordagem segmentada com o objetivo de facilitar a sua manutenção. A nova proposta consiste em modelar a parte estacionária separada da parte dinâmica, diferentemente do que é feito na metodologia tradicional, onde o modelo dinâmico é gerado diretamente dos dados de processo. O modelo estacionário é obtido através de uma regressão PLS (Partial Least Squares), enquanto as dinâmicas são inseridas posteriormente utilizando-se um algoritmo de otimização. A técnica é aplicada a uma coluna de destilação e o resultado obtido é semelhante ao de inferências dinâmicas e estáticas desenvolvidas com métodos tradicionais. Outras etapas do desenvolvimento de inferências também são investigadas. Na seleção de variáveis, métodos estatísticos são comparados com a busca exaustiva e se conclui este último deve ser usado como padrão, visto que custo computacional não é mais um problema. Também são apresentadas boas práticas no pré-tratamento de dados, remoção do tempo morto do cromatógrafo modelado e detecção de estados estacionários. / Soft-sensors have several applications in the chemical processes industry and are essential for the success of advanced control projects. Its performance will always be linked to the performance of the soft-sensor, so it is important to maintain its quality over time. In this paper, a new methodology is suggested for the development of data-based soft-sensors following a segmented approach in order to facilitate its maintenance. The new proposal is to model the stationary part separated from the dynamic, unlike the traditional methodology where the dynamic model is generated directly from process data. The stationary model is obtained by a PLS (Partial Least Squares) regression, while the dynamics are inserted using an optimization algorithm. The technique is applied to a distillation column and its performance is similar to dynamic and static soft-sensors developed using traditional methods. Other steps in the development of soft-sensors are also investigated. In variable selection issue, statistical methods are compared with the testing of all possibilities; the latter should be used as default, since computational cost is no longer a problem. We also present best practices in data pre-processing, gas chromatograph dead-time removal and steady state detection.