No description available.

No description available.

Orientador: Roberto Wagner Lourenço / Resumo: O controle dos vetores do mosquito Aedes aegypti apresenta como um dos principais problemas de saúde pública no mundo representado por duas vertentes importantes como o cuidado com o meio ambiente e pelo investimento em ciência. A preocupação com o meio ambiente é uma das maneiras mais eficazes de conter a reprodução do mosquito Aedes aegypti que, segundo Organização Mundial da Saúde, infecta cerca de 100 milhões de pessoas por ano. Esta pesquisa teve como objetivo principal analisar a distribuição espacial dos casos de dengue confirmados autóctones e importados, bem como dos recipientes de larvas do mosquito no município de Itu, São Paulo no período de 2005 a 2014. Para o desenvolvimento do trabalho foi realizada uma pesquisa junto à Vigilância Epidemiológico da área de estudo para obtenção dos dados, acompanhada de pesquisa bibliográfica e trabalho de campo. Os dados foram tratados por meio de técnicas de geoprocessamento e análise estatística exploratória. Os resultados obtidos demonstraram uma crescente alta de casos e de incidência ao longo dos anos estudados da faixa de 10 anos, sendo que os casos autóctones se sobrepõe em termos de ocorrência aos importados em 88 %. Além disso, foi possível verificar uma distribuição espacial com maiores ocorrências nas regiões centrais da área urbana do município estudado, apresentando as maiores concentrações nos anos de 2007, 2009, 20011 e 2013 do total analisado, com distribuição tendendo para a direção sudeste-noroeste da área de ... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: The control of the vectors of the mosquito Aedes aegypti presents as one of the main problems of public health in the world acted by two important slopes as the care with the environment and for the investment in science. The concern with the environment is one in the most effective ways of containing the reproduction of the mosquito Aedes aegypti that, second World Organization of the Health, infects about 100 million people a year. This research had as main objective to analyze the space distribution of the cases of primness confirmed autochthonous and mattered, as well as of the containers of larvas of the mosquito in the municipal district of Itu, São Paulo in the period of 2005 the 2014. For the development of the work a research was accomplished the Surveillance Epidemic of the study area close to for obtaining of the data, accompanied of bibliographical research and field work. The data were treated through geoprocessamento techniques and exploratory statistical analysis. The obtained results demonstrated a crescent high of cases and of incidence along the studied years of the 10 year-old strip, and the autochthonous cases are put upon in occurrence terms to the mattered in 88%. Besides, it was possible to verify a space distribution with larger occurrences in the central areas of the urban area of the studied municipal district, presenting the largest concentrations in the years of 2007, 2009, 20011 and 2013 of the analyzed total, with distribution tending for the sou... (Complete abstract click electronic access below) / Mestre

The purpose of the present thesis is to show that hyperspectral data can be used for discrimination between different tree species. The data set used in this study contains the hyperspectral measurements of leaves of seven savannah tree species. The data is high-dimensional and shows large within-class variability combined with small between-class variability which makes discrimination between the classes challenging. We employ two classification methods: G-nearest neighbour and feed-forward neural networks. For both methods, direct 7-class prediction results in high misclassification rates. However, binary classification works better. We constructed binary classifiers for all possible binary classification problems and combine them with Error Correcting Output Codes. We show especially that the use of 1-nearest neighbour binary classifiers results in no improvement compared to a direct 1-nearest neighbour 7-class predictor. In contrast to this negative result, the use of neural networks binary classifiers improves accuracy by 10% compared to a direct neural networks 7-class predictor, and error rates become acceptable. This can be further improved by choosing only suitable binary classifiers for combination.

No description available.

This thesis discusses four novel statistical methods and approximate inference techniques for analyzing structured neural and molecular sequence data. The main contributions are new algorithms for approximate inference and learning in Bayesian latent variable models involving spatial statistics and nonlinear dynamics. First, we propose an amortized variational inference method to separate a set of overlapping signals into spatially localized source functions without knowledge of the original signals or the mixing process. In the second part of this dissertation, we discuss two approaches for uncovering nonlinear, smooth latent dynamics from sequential data. Both algorithms construct variational families on extensions of nonlinear state space models where the underlying systems are described by hidden stochastic differential equations. The first method proposes a structured approximate posterior describing spatially-dependent linear dynamics, as well as an algorithm that relies on the fixed-point iteration method to achieve convergence. The second method proposes a variational backward simulation technique from an unbiased estimate of the marginal likelihood defined through a subsampling process. In the final chapter, we develop connections between discrete and continuous variational sequential search for Bayesian phylogenetic inference. We propose a technique that uses sequential search to construct a variational objective defined on the composite space of non-clock phylogenetic trees. Each of these techniques are motivated by real problems within computational biology and applied to provide insights into the underlying structure of complex data.

Thesis (Ph.D. (Statistics)) -- University of Limpopo, 2021 / In spatial statistics, several methods have been developed to measure the extent of local and global spatial dependence (clustering) in measured data across areas in a region of research interest. These methods are now routinely implemented in most Geographical Information Systems (GIS) and statistical computer packages. However, spatial statistics for measuring joint spatial dependence of multiple spatial measurement and outcome data have not been well developed. A naive analysis would simply apply univariate spatial dependence methods to each data separately. Though this is simple and straightforward, it ignores possible relationships between multiple spatial data because they may be measuring the same phenomena. Limited work has been done on extending the Moran’s index, a commonly used and applied univariate measure of spatial clustering, to bivariate Moran’s index in order to assess spatial dependence for two spatial data. The overall aim of this PhD was to develop multivariate spatial clustering methods for multiple spatial data, especially in the health sciences. Our proposed multivariate spatial clustering statistic is based on the fundamental theory regarding canonical correlations. We firstly reviewed and applied univariate and bivariate Moran’s indexes to spatial analyses of multiple non-communicable diseases and related risk factors in South Africa. Then we derived our proposed multivariate spatial clustering method, which was evaluated by simulation studies and applied to a spatial analysis of multiple non-communicable diseases and related risk factors in South Africa. Simulation studies showed that our proposed multivariate spatial statistic was able to identify correctly clusters of areas with high risks as well as clusters with low risk.

This work presents methods used to obtain geometric models of spatial mechanisms which can be realized in hardware. Each model is created automatically from the kinematic description of a mechanism. The models are tested for interference between joints and links. Models with interfering links or joints are reshaped automatically into an interference-free configuration. An investigation of the relative efficiency of different interference detection techniques is discussed. A method for determining interferences based on vector loop equations was developed for this work. Other approaches for interference detection include parametric space and a method using parallel coordinates. 2000 line segments were randomly generated to test the three methods. No significant difference between the three techniques was found, but a coarse detection scheme was developed based on observations of intersection conditions in parallel coordinates. The coarse detection technique reduced interference detection times by 48%. The concept of joint positioning freedoms is presented formally for the first time. Using a unidirectional avoidance strategy along a straight line, these repositioning freedoms are exploited in a manner which guarantees the elimination of interferences for revolute, prismatic, and cylindric joints. A unique method for optimal orientation of spheric joint ball-cup pairs is described. Points from an inverse image of the attachment piece for the ball are mapped onto a unit sphere in the reference frame of the cup. The axis of a bounding cone is then used to align the attachment piece for the cup. The method minimizes the chances for collisions between the cup and the ball attachment piece. Elements which attach the joints are modeled as three segments. This has proven to be an optimal representation. Interferences with these elements are eliminated using the elliptical projection of circular paths onto a plane which is perpendicular to the axis of symmetry for an intruding object. Several examples are given illustrating the successful generation of interference-free spatial mechanism models. The mechanisms include an RSSR, an RPCS, an RCCC, and an RRRRRRR. / Ph. D.

For robots to move out of the lab and into the real-world, they must be able to plan routes not only through space but through time as well. The introduction of a time factor to the planning process implies that robots must reason about other processes and agents that move through space independently of the robot's actions. This thesis presents an integrated route planner and spatial representation system for planning real-time paths through dynamic domains called Robonav. Robonav will find the safest 9 most efficient route through time and space as described by an evaluation function. Due to the design of the spatial representation and the mechanics of the algorithm, Robonav has an isomorphic mapping onto a machine with a highly parallel SIMD architecture. When Robonav is operated in a predictable domain, paths are found in O(p) time (where p is the length of a path). In unpredictable domains, where Robonav is operated in incremental mode, paths are found and executed in O(p²) time. / Master of Science

No description available.