The Quinato wetland, a remnant of a Pleistocene river course through northeastern Bolivia, has undoubtedly been the site of human landscape modification and domestication by pre-Columbian peoples. A 2021 study suggests that these modification practices, which have been tied to seasonal adaptation, were quite different between areas of the wetland. In response to these findings, the present study uses unsupervised classifications from the 50-year span of existent Landsat satellite imagery data, dating from 1972 to 2022, to create a chronological profile of the wetland. This record allows for the assessment of how yearly, seasonal changes to wetland growth and shrinkage contribute to longer-term trends. No significant increase or decrease in wetland size overall is suggested by the data, although a distinct, common seasonal pattern is detectable. This data, narrowed to the two sites investigated in the previously mentioned study, shows a similar seasonal patterning as that of the larger wetland at these two sites, but also detects a greater stability in wetland area for the region that was first modified by pre-Columbian peoples.

A Unidade de Gerenciamento de Recursos Hídricos n° 11 (UGRHI 11) correspondente à Bacia Hidrográfica do Rio Ribeira de Iguape e Litoral Sul e pequenas bacias litorâneas adjacentes, se localiza ao sul do Estado de São Paulo. Esta unidade apresenta grande diversidade de ambientes terrestres e aquáticos, envolvendo extensas áreas de relevo serrano, com fortes declividades e várzeas encaixadas e um setor composto por planícies costeiras, manguezais, terraços marinhos e fluviais. É uma região sensível não só nos aspectos da biodiversidade, mas também no âmbito dos processos geomorfológicos, pois se trata de ambiente extremamente propício aos desencadeamentos de processos erosivos de todos os tipos, bem como altamente susceptível a escorregamentos de encostas e rolamentos de blocos. O objetivo deste trabalho foi desenvolver possíveis cenários futuros de uso e ocupação do solo, baseando-se na evolução deste uso durante 24 anos (1986 - 2010) e prever cenários para 2025, analisando o crescimento ou regressão das classes de uso e ocupação do solo e considerando nessa evolução temporal a ação de agentes modificadores do terreno. A primeira fase da metodologia foi composta pela elaboração de mapas de suscetibilidade a ero são e movimentos de massa utilizando os métodos RUSLE e Combinação Linear Ponderada, respectivamente. Posteriormente, através de classificação não supervisionada , elaborou-se mapas de uso e ocupação do solo da UGRHI-11 referente a 1986, 1999 e 2010. Através da aplicação das técnicas de Titus e Narayanan, Pfeffer, Ramhstorf e modelo do Intergovernmental Panel on Climate Change, no ArcGIS, calculou-se a vulnerabilidade do Complexo Estuarino-Lagunar de Cananéia-Iguape a um potencial aumento do nível do mar em 2025, 2050 e 2100. Os mapas de uso e ocupação do solo de 1986, 1999 e 2010, aliados aos mapas de suscetibilidade à erosão e movimentos de massa foram utilizados nas Cadeias de Markov, acopladas a um algoritmo de Autômato Celular, no IDRISI Taiga, para a simulação do uso e ocupação do solo da UGRHI-11, em 2025. Na última fase da metodologia, foram avaliadas as classes de uso e ocupação do solo que possivelmente serão afetadas por um evento de máxima preamar, em 2025. Como conclusão, a utilização de diferentes tipos de ferramentas das geotecno logias mostrou-se eficaz e mesmo integrando diferentes \"frentes\" de trabalho: mapeamento de áreas suscetíveis a movimentos de massa e erosão; mapeamento do uso e cobertura do solo da UGRHI-11 e simulação do cenário de 2025; além do cálculo da vulnerabilidade a um potencial aumento do nível do mar (2025, 2050 e 2100). Foi possível agregar todos estes produtos e elaborar o produto final, a quantificação das áreas das classes de uso e ocupação do solo do Complexo Estuarino-Lagunar de Iguape-Cananéia, em 2025. / The water resources managing unit 11 (UGRHI-11) corresponds to Ribeira de Iguape r iver Drainage Basin and São Paulo State Southern Coast and small adjacent basins is located in south of São Paulo State. This unit presents great diversity of terrestrial and aquatic environments, big areas of dissected relief with high slopes and lowland and a sector consists of coastal plains, mangroves, marine and river terraces. It is a region sensible not only to biodiversity aspects but included to geomorphology process because it\'s an environment conducive to erosion and landslides processes. The goal of this search was develop possible scenarios of soil use and occupation based in the evolution during 24 years (1986 - 2010) and forecast scenarios to 2025, analyzing the growth or retraction of classes of soil use and occupation and considering in this temporal evolution the actions of terrain modifiers. In the first step of methodology was developed the maps of susceptibility to erosion and landslides with the use of RUSLE and Weight Linear Combination. In the next step by using unsupervised classification were prepared maps of soil use and occupation of UGRHI -11 relative to years 1986, 1999 and 2010. The methods of Titus y Narayanan, Pfeffer, Ramhstorf and Intergovernmental Panel on Climate Change model were processed in the ArcGIS 10 program with the goal of calculate the vulnerability of Complexo Estuarino -Lagunar de Cananéia-Iguape to a potential increase in sea level in 2025, 2050 and 2100. The maps of soil use and occupation of 1986, 1999 and 2010 allies to the maps of erosion and landslides were used in Markov Chain Analysis and Cellular Automata in the IDRISI Taiga to simulation of classes of soil use and occupation of UGRHI-11 in 2025. In the last step were evaluated the classes of soil use and occupation that will be affected by one possible high tide event in 2025. The conclusion is that the application of differents kinds of geotecnology tools was effective even integrating different jobs: mapping of susceptible areas to erosion and landslides; mapping of soil use and simulation of scenarios o f 2025; beyond the calculation of vulnerability to a potential increase in sea level in (2025, 2050 and 2100). It was possible to aggregate all these products and to elaborate the last product the quantification of areas of classes of soil use and occupation of Complexo Estuarino-Lagunar de Iguape-Cananéia, in 2025.

A Unidade de Gerenciamento de Recursos Hídricos n° 11 (UGRHI 11) correspondente à Bacia Hidrográfica do Rio Ribeira de Iguape e Litoral Sul e pequenas bacias litorâneas adjacentes, se localiza ao sul do Estado de São Paulo. Esta unidade apresenta grande diversidade de ambientes terrestres e aquáticos, envolvendo extensas áreas de relevo serrano, com fortes declividades e várzeas encaixadas e um setor composto por planícies costeiras, manguezais, terraços marinhos e fluviais. É uma região sensível não só nos aspectos da biodiversidade, mas também no âmbito dos processos geomorfológicos, pois se trata de ambiente extremamente propício aos desencadeamentos de processos erosivos de todos os tipos, bem como altamente susceptível a escorregamentos de encostas e rolamentos de blocos. O objetivo deste trabalho foi desenvolver possíveis cenários futuros de uso e ocupação do solo, baseando-se na evolução deste uso durante 24 anos (1986 - 2010) e prever cenários para 2025, analisando o crescimento ou regressão das classes de uso e ocupação do solo e considerando nessa evolução temporal a ação de agentes modificadores do terreno. A primeira fase da metodologia foi composta pela elaboração de mapas de suscetibilidade a ero são e movimentos de massa utilizando os métodos RUSLE e Combinação Linear Ponderada, respectivamente. Posteriormente, através de classificação não supervisionada , elaborou-se mapas de uso e ocupação do solo da UGRHI-11 referente a 1986, 1999 e 2010. Através da aplicação das técnicas de Titus e Narayanan, Pfeffer, Ramhstorf e modelo do Intergovernmental Panel on Climate Change, no ArcGIS, calculou-se a vulnerabilidade do Complexo Estuarino-Lagunar de Cananéia-Iguape a um potencial aumento do nível do mar em 2025, 2050 e 2100. Os mapas de uso e ocupação do solo de 1986, 1999 e 2010, aliados aos mapas de suscetibilidade à erosão e movimentos de massa foram utilizados nas Cadeias de Markov, acopladas a um algoritmo de Autômato Celular, no IDRISI Taiga, para a simulação do uso e ocupação do solo da UGRHI-11, em 2025. Na última fase da metodologia, foram avaliadas as classes de uso e ocupação do solo que possivelmente serão afetadas por um evento de máxima preamar, em 2025. Como conclusão, a utilização de diferentes tipos de ferramentas das geotecno logias mostrou-se eficaz e mesmo integrando diferentes \"frentes\" de trabalho: mapeamento de áreas suscetíveis a movimentos de massa e erosão; mapeamento do uso e cobertura do solo da UGRHI-11 e simulação do cenário de 2025; além do cálculo da vulnerabilidade a um potencial aumento do nível do mar (2025, 2050 e 2100). Foi possível agregar todos estes produtos e elaborar o produto final, a quantificação das áreas das classes de uso e ocupação do solo do Complexo Estuarino-Lagunar de Iguape-Cananéia, em 2025. / The water resources managing unit 11 (UGRHI-11) corresponds to Ribeira de Iguape r iver Drainage Basin and São Paulo State Southern Coast and small adjacent basins is located in south of São Paulo State. This unit presents great diversity of terrestrial and aquatic environments, big areas of dissected relief with high slopes and lowland and a sector consists of coastal plains, mangroves, marine and river terraces. It is a region sensible not only to biodiversity aspects but included to geomorphology process because it\'s an environment conducive to erosion and landslides processes. The goal of this search was develop possible scenarios of soil use and occupation based in the evolution during 24 years (1986 - 2010) and forecast scenarios to 2025, analyzing the growth or retraction of classes of soil use and occupation and considering in this temporal evolution the actions of terrain modifiers. In the first step of methodology was developed the maps of susceptibility to erosion and landslides with the use of RUSLE and Weight Linear Combination. In the next step by using unsupervised classification were prepared maps of soil use and occupation of UGRHI -11 relative to years 1986, 1999 and 2010. The methods of Titus y Narayanan, Pfeffer, Ramhstorf and Intergovernmental Panel on Climate Change model were processed in the ArcGIS 10 program with the goal of calculate the vulnerability of Complexo Estuarino -Lagunar de Cananéia-Iguape to a potential increase in sea level in 2025, 2050 and 2100. The maps of soil use and occupation of 1986, 1999 and 2010 allies to the maps of erosion and landslides were used in Markov Chain Analysis and Cellular Automata in the IDRISI Taiga to simulation of classes of soil use and occupation of UGRHI-11 in 2025. In the last step were evaluated the classes of soil use and occupation that will be affected by one possible high tide event in 2025. The conclusion is that the application of differents kinds of geotecnology tools was effective even integrating different jobs: mapping of susceptible areas to erosion and landslides; mapping of soil use and simulation of scenarios o f 2025; beyond the calculation of vulnerability to a potential increase in sea level in (2025, 2050 and 2100). It was possible to aggregate all these products and to elaborate the last product the quantification of areas of classes of soil use and occupation of Complexo Estuarino-Lagunar de Iguape-Cananéia, in 2025.

Le traitement de nuages de points 3D de lidars permet aux robots mobiles autonomes terrestres de construire des modèles sémantiques de l'environnement extérieur dans lequel ils évoluent. Ces modèles sont intéressants car ils représentent des informations qualitatives, et ainsi donnent à un robot la capacité de raisonner à un niveau plus élevé d'abstraction. Le coeur d'un système de modélisation sémantique est la capacité de classifier les observations venant du capteur. Nous proposons un système de classification centré sur l'apprentissage non-supervisé. La prémière couche, la couche intermédiaire, consiste en un modèle de mélange gaussien. Ce modèle est déterminé de manière non-supervisée lors d'une étape de training. Il definit un ensemble de classes intermédiaires qui correspond à une partition fine des classes présentes dans l'environnement. La deuxième couche, la couche finale, consiste en un regroupement des classes intermédiaires dans un ensemble de classes finales qui, elles, sont interprétables dans le contexte de la tâche ciblée. Le regroupement est déterminé par un expert lors de l'étape de training, de manière supervisée, mais guidée par les classes intermédiaires. L'évaluation est basée sur deux jeux de données acquis avec de différents lidars et possédant différentes caractéristiques. L'évaluation est quantitative pour l'un des jeux de données, et qualitative pour l'autre. La concéption du système utilise la procédure standard de l'apprentissage, basée sur les étapes de training, validation et test. L'opération suit la pipeline standard de classification. Le système est simple, et ne requiert aucun pré-traitement ou post-traitement. / The processing of 3D lidar point clouds enable terrestrial autonomous mobile robots to build semantic models of the outdoor environments in which they operate. Such models are interesting because they encode qualitative information, and thus provide to a robot the ability to reason at a higher level of abstraction. At the core of a semantic modelling system, lies the capacity to classify the sensor observations. We propose a two-layer classi- fication model which strongly relies on unsupervised learning. The first, intermediary layer consists of a Gaussian mixture model. This model is determined in a training step in an unsupervised manner, and defines a set of intermediary classes which is a fine-partitioned representation of the environment. The second, final layer consists of a grouping of the intermediary classes into final classes that are interpretable in a considered target task. This grouping is determined by an expert during the training step, in a process which is supervised, yet guided by the intermediary classes. The evaluation is done for two datasets acquired with different lidars and possessing different characteristics. It is done quantitatively using one of the datasets, and qualitatively using another. The system is designed following the standard learning procedure, based on a training, a validation and a test steps. The operation follows a standard classification pipeline. The system is simple, with no requirement of pre-processing or post-processing stages.

In this dissertation, we extend several relatively new developments in statistical model selection and data mining in order to improve one of the workhorse statistical tools - mixture modeling (Pearson, 1894). The traditional mixture model assumes data comes from several populations of Gaussian distributions. Thus, what remains is to determine how many distributions, their population parameters, and the mixing proportions. However, real data often do not fit the restrictions of normality very well. It is likely that data from a single population exhibiting either asymmetrical or nonnormal tail behavior could be erroneously modeled as two populations, resulting in suboptimal decisions. To avoid these pitfalls, we develop the mixture model under a broader distributional assumption by fitting a group of multivariate elliptically-contoured distributions (Anderson and Fang, 1990; Fang et al., 1990). Special cases include the multivariate Gaussian and power exponential distributions, as well as the multivariate generalization of the Student’s T. This gives us the flexibility to model nonnormal tail and peak behavior, though the symmetry restriction still exists. The literature has many examples of research generalizing the Gaussian mixture model to other distributions (Farrell and Mersereau, 2004; Hasselblad, 1966; John, 1970a), but our effort is more general. Further, we generalize the mixture model to be non-parametric, by developing two types of kernel mixture model. First, we generalize the mixture model to use the truly multivariate kernel density estimators (Wand and Jones, 1995). Additionally, we develop the power exponential product kernel mixture model, which allows the density to adjust to the shape of each dimension independently. Because kernel density estimators enforce no functional form, both of these methods can adapt to nonnormal asymmetric, kurtotic, and tail characteristics. Over the past two decades or so, evolutionary algorithms have grown in popularity, as they have provided encouraging results in a variety of optimization problems. Several authors have applied the genetic algorithm - a subset of evolutionary algorithms - to mixture modeling, including Bhuyan et al. (1991), Krishna and Murty (1999), and Wicker (2006). These procedures have the benefit that they bypass computational issues that plague the traditional methods. We extend these initialization and optimization methods by combining them with our updated mixture models. Additionally, we “borrow” results from robust estimation theory (Ledoit and Wolf, 2003; Shurygin, 1983; Thomaz, 2004) in order to data-adaptively regularize population covariance matrices. Numerical instability of the covariance matrix can be a significant problem for mixture modeling, since estimation is typically done on a relatively small subset of the observations. We likewise extend various information criteria (Akaike, 1973; Bozdogan, 1994b; Schwarz, 1978) to the elliptically-contoured and kernel mixture models. Information criteria guide model selection and estimation based on various approximations to the Kullback-Liebler divergence. Following Bozdogan (1994a), we use these tools to sequentially select the best mixture model, select the best subset of variables, and detect influential observations - all without making any subjective decisions. Over the course of this research, we developed a full-featured Matlab toolbox (M3) which implements all the new developments in mixture modeling presented in this dissertation. We show results on both simulated and real world datasets. Keywords: mixture modeling, nonparametric estimation, subset selection, influence detection, evidence-based medical diagnostics, unsupervised classification, robust estimation.

Submitted by JOSELINA DE SOUZA CORREIA null (joselina.scorreia@gmail.com) on 2017-01-10T13:04:37Z No. of bitstreams: 1 Tese oficial _ Joselina de Souza Correia . 2017.pdf: 2214453 bytes, checksum: 2a09bae07d69c106a4dd8c34d57ecd39 (MD5) / Approved for entry into archive by Juliano Benedito Ferreira (julianoferreira@reitoria.unesp.br) on 2017-01-12T13:56:01Z (GMT) No. of bitstreams: 1 correia_js_dr_bot.pdf: 2214453 bytes, checksum: 2a09bae07d69c106a4dd8c34d57ecd39 (MD5) / Made available in DSpace on 2017-01-12T13:56:01Z (GMT). No. of bitstreams: 1 correia_js_dr_bot.pdf: 2214453 bytes, checksum: 2a09bae07d69c106a4dd8c34d57ecd39 (MD5) Previous issue date: 2016-11-18 / Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) / Diante das situações de escassez hídrica que ocorrem no Semiárido Brasileiro, e do contínuo aumento populacional naquela região são procurados meios que otimizem o uso da água. O objetivo desse trabalho foi analisar e criar uma perspectiva, dentro das variações ocorridas nos padrões de uso e ocupação do solo, por meio de imagens de satélites e da infiltração de água no solo, para subsidiar o uso racional da área, da água no Perímetro de Irrigação Pontal Sul, em Pernambuco. Para isso, foi feito o uso de geotecnologias, para fornecer elementos subsidiando uma melhor gestão em larga escala. Duas imagens LANDSAT para os anos2000-2015, foram analisadas e classificadas por meio do programa ArcGis, para comparação dos diferentes tipos de uso do solo. As classes e subclasses foram definidas por processo não supervisionado, seguido por supervisionado, sendo o perímetro dividido em duas classes, área antropizada e área natural, e esses subdivididos em solo descoberto e agricultura irrigada, e em Caatinga densa e Caatinga raleada, respectivamente. As mesmas imagens foram analisadas pelo programa IDRISI, e seus resultados confrontados com os encontrados pelo ArcGis. Foram utilizados dados de 21 testes de infiltração, da água no solo determinados por meio da metodologia dos anéis concêntricos, de onde se obteve a velocidade de infiltração básica (VIB). Os dados encontrados foram ajustados por vários modelos, e o de Kostiakov e a função potência foram os mais qualificados para os tipos de solo do perímetro. Quanto ao uso da terra, a Caatinga densa está presente em 22,1% e 24,8%; a Caatinga rala representa 51,2% e 39,1%; e a classe solo está descoberto em 26,75% e 33,64% da área do perímetro, respectivamente em 2000 e 2015. A agricultura irrigada foi visível apenas no ano de 2015 com 2,45% da área. Quando comparados, os programas evidenciaram algumas similaridades, e relativa coerência no uso e transformação das classes do solo, sendo que o IDRISI atribuiu a algumas classes uma extensão superior ao determinado pelo ArcGis. Como divergência, o programa IDRISI registrou presença de água superficial no primeiro cenário e agricultura irrigada em ambos os anos. A VIB foi classificada como muito alta e aponta a região como arenosa sob as condições analisadas, o que dificulta sua retenção hídrica e favorece sua infiltração. Os cenários sinalizam regiões parcialmente similares como zonas potenciais, para expansão da agricultura irrigada, legitimando a eficiência dessas geotecnologias na gestão e perspectiva de uso do perímetro de irrigação. / In face of the water stress situations that occur in the Brazilian semi-arid, as well as the continued population growth in that region, the use of water resources should be maximized. Thus, the aim of this study was to analyze and create a perspective within the variations of land use patterns, through the image analysis and soil water infiltration, for guiding the use of irrigated area inside of the Pontal Sul Irrigation Scheme, state of Pernambuco, Brazil. For this, the use of geotechnology was made to provide elements for a better area management. Satellite images (2000-2015), provided by the INPE and Codevasf, were analyzed and classified by the ArcGIS software for investigation and comparison of land use. The classes and subclasses were defined by unsupervised process, followed by supervised one, and the irrigation scheme was divided into two classes, anthropic area and natural area, and these were subdivided into discovered and irrigated agriculture soil, and dense Caatinga and thinned Caatinga, respectively. The same images were analyzed by IDRISI software, and the results compared with those found by ArcGis. Data from 21 infiltration tests were used, determined by the methodology of concentric rings, where basic infiltration rate (BIR) was derived. The data were adjusted for several models, and the Kostiakov and the power function models were the most qualified to the irrigation scheme soils. Regarding the use of the land, dense Caatinga is present in 22.1% and 24.8%; the thinned Caatinga represents 51.2% and 39.1%; and the soil is uncovered in 26.75% and 33.64% of the irrigation scheme area, respectively in 2000 and 2015. Irrigated agriculture was visible only in the year 2015, in 2.45% of the area. Both software showed some similarities and consistency on the use and transformation of soil classes, and IDRISI assigned to a higher extent class determined by the ArcGis. As divergence IDRISI software recorded the presence of surface water in the first scenario and irrigated agriculture in both years. The BIR was classified as very high and indicated that soils as sandy under the conditions analyzed, which makes its water retention low and favors water infiltration. The scenarios indicate partially similar regions as potential areas for expansion of irrigated agriculture, legitimizing the efficiency of these geotechnology in management and in the perspective of the irrigation scheme use. / CNPq:161122/2012-4

Au vu des changements globaux actuels engendrés en grande partie par l'être humain, il devient nécessaire de comprendre les moteurs de la stabilité des communautés d'êtres vivants. La synchronie des séries temporelles d'abondances fait partie des mécanismes les plus importants. Cette thèse propose trois angles différents permettant de répondre à différentes questions en lien avec la synchronie interspécifique ou spatiale. Les travaux présentés trouvent des applications en dehors du cadre écologique. Un premier chapitre est consacré à l'estimation du rang effectif de matrices à valeurs dans ℝ ou ℂ. Nous apportons ainsi des outils permettant de mesurer le taux de synchronisation d'une matrice d'observations. Dans le deuxième chapitre, nous nous basons sur les travaux existants sur le problème de détection de ruptures sur les chaînes afin de proposer plusieurs algorithmes permettant d'adapter ce problème au cas des arbres. Les méthodes présentées peuvent être utilisées sur la plupart des données nécessitant d'être représentées sous la forme d'un arbre. Afin d'étudier les liens entre la synchronie interspécifique et les tendances à long termes ou les traits d'espèces de papillons, nous proposons dans le dernier chapitre d'adapter des méthodes de clustering et d'apprentissage supervisé comme les Random Forest ou les Réseaux de Neurones artificiels à des données écologiques. / In the view of actual global changes widely caused by human activities, it becomes urgent to understand the drivers of communities' stability. Synchrony between time series of abundances is one of the most important mechanisms. This thesis offers three different angles in order to answer different questions linked to interspecific and spatial synchrony. The works presented find applications beyond the ecological frame. A first chapter is dedicated to the estimation of effective rank of matrices in ℝ or ℂ. We offer tools allowing to measure the synchronisation rate of observations matrices. In the second chapter, we base on the existing work on change-points detection problem on chains in order to offer algorithms which detects change-points on trees. The methods can be used with most data that have to be represented as a tree. In order to study the link between interspecific synchrony and long term tendencies or traits of butterflies species, we offer in the last chapter adaptation of clustering and supervised machine learning methods, such as Random Forest or Artificial Neural Networks to ecological data.

This dissertation develops new techniques employing the Low-rank and Sparse Representation approaches to improve the performance of state-of-the-art algorithms in hyperspectral image analysis. The contributions of this dissertation are outlined as follows. 1) Low-rank and sparse representation approaches, i.e., low-rank representation (LRR) and low-rank subspace representation (LRSR), are proposed for hyperspectral image analysis, including target and anomaly detection, estimation of the number of signal subspaces, supervised and unsupervised classification. 2) In supervised target and unsupervised anomaly detection, the performance can be improved by using the LRR sparse matrix. To further increase detection accuracy, data is partitioned into several highly-correlated groups. Target detection is performed in each group, and the final result is generated from the fusion of the output of each detector. 3) In the estimation of the number of signal subspaces, the LRSR low-rank matrix is used in conjunction with direct rank calculation and soft-thresholding. Compared to the state-of-the-art algorithms, the LRSR approach delivers the most accurate and consistent results across different datasets. 4) In supervised and unsupervised classification, the use of LRR and LRSR low-rank matrices can improve classification accuracy where the improvement of the latter is more significant. The investigation on state-of-the-art classifiers demonstrate that, as a pre-preprocessing step, the LRR and LRSR produce low-rank matrices with fewer outliers or trivial spectral variations, thereby enhancing class separability.

Mathematical models and their associated computer simulations are nowadays widely used in several research fields, such as natural sciences, engineering, as well as social sciences. In the context of systems biology, they provide a rigorous way to investigate how complex regulatory pathways are connected and how the disruption of these processes may contribute to the develop- ment of a disease, ultimately investigating the suitability of specific molecules as novel therapeutic targets. In the last decade, the launching of the precision medicine initiative has motivated the necessity to define innovative computational techniques that could be used for customizing therapies. In this context, the combination of mathematical models and computer strategies is an essential tool for biologists, which can analyze complex system pathways, as well as for the pharmaceutical industry, which is involved in promoting programs for drug discovery. In this dissertation, we explore different modeling techniques that are used for the simulation and the analysis of complex biological systems. We analyze the state of the art for simulation algorithms both in the stochastic and in the deterministic frameworks. The same dichotomy has been studied in the context of sensitivity analysis, identifying the main pros and cons of the two approaches. Moreover, we studied the quantitative system pharmacology (QSP) modeling approach that elucidates the mechanism of action of a drug on the biological processes underlying a disease. Specifically, we present the definition, calibration and validation of a QSP model describing Gaucher disease type 1 (GD1), one of the most common lysosome storage rare disorders. All of these techniques are finally combined to define a novel computational pipeline for patient stratification. Our approach uses modeling techniques, such as model simulations, sensitivity analysis and QSP modeling, in combination with experimental data to identify the key mechanisms responsible for the stratification. The pipeline has been applied to three test cases in different biological contexts: a whole-body model of dyslipidemia, the QSP model of GD1 and a QSP model of cardiac electrophysiology. In these test cases, the pipeline proved to be accurate and robust, allowing the interpretation of the mechanistic differences underlying the phenotype classification.

Accelerometers deployed within smart structures produce a wealth of vibration data that can be analyzed to infer information about the types of acceleration events that are occurring within the structure. In the case of monitored smart buildings, some of these acceleration events are linked to occupant behavior, such as walking, operating machinery, closing doors, etc. The identification and classification of such events has many potential applications within a smart structure or city. Understanding occupant patterns could be beneficial for operations, retail, or HVAC management, as it could be used to monitor occupancy flow with a relatively sparse sensor network. It may also have detrimental implications in terms of cybersecurity, where such information could be mined for malicious practices if unauthorized access to the data was obtained. This work presents methods for the detection and classification of vibration events in an experimental smart building, Goodwin Hall at Virginia Tech. Goodwin Hall's 200+ accelerometer network is used to gather acceleration data, from which vibration events are automatically detected and clustered. The presence of a vibration event is detected from a raw acceleration signal with an adaptive RMS threshold method. A feature vector is then created for each extracted event as areas under regions of the FFT of the event's acceleration signal. The feature vectors are then mapped into a low-dimensional space using principal component analysis, where they are clustered with various unsupervised algorithms. These processes have shown to be successful when gathering vibration events from a single-sensor setup, but pose challenges when expanded to a multi-sensor network. Because of this, expanded applications such as a semi-supervised classifier for events detected anywhere in the building are currently still under development. This semi-supervised process, combined with the known location of each sensor would allow inferences to be drawn about the frequency of different activity types in regions of the building not captured in the labeled data. Future work intends to address these multi-sensor challenges with adjustments to the algorithm process. / Master of Science / All objects experience vibrations when they are disturbed by some force. In the case of this work, the object is complex, a classroom building, but the principle still stands. When the building is disturbed by a force it will vibrate, even if the force is small, such as a person walking down a hallway or closing a door. The vibrations caused by these 'events' are unique to the type of event, that is, footstep vibrations will be different from door vibrations. These vibrations are observed with accelerometers, and the corresponding signal is used to determine what type of event caused the vibration. First, an event is automatically detected within the signal and separated from it. Second, characteristics unique to the signal are identified, a process known as 'feature extraction.' Finally, those features are used to distinguish the event from others and to identify what had caused it based on previous experimental data. The ability to detect these events and classify them introduces many interesting applications, including any that would stem from occupant detection, including improved security or operations, retail, or HVAC management. The methods here may also be applicable to other applications, such as monitoring bridges and machinery, or for developing cutting-edge smartphone applications with the accelerometer that is built in.