1000 resultados para Probabilidade de classificação errada
Resumo:
O problema de classificar pessoas de acordo com o tamanho da induração, na prova tuberculínica segundo a técnica de Mantoux, é resolvido, para um conjunto de dados obtidos em uma população genérica, utilizando-se o critério estatístico de "melhores regiões possíveis de classificação". São obtidas estimativas das probabilidades de classificação errada.
Resumo:
Pós-graduação em Agronomia (Energia na Agricultura) - FCA
Resumo:
Pós-graduação em Agronomia (Energia na Agricultura) - FCA
Resumo:
Dissertação de Mestrado, Engenharia Eletrónica e Telecomunicações, Faculdade de Ciências e Tecnologia, Universidade do Algarve, 2014
Resumo:
Relatório de projecto apresentado como requisito parcial para obtenção do grau de mestre em Estatística e Gestão de Informação.
Resumo:
A aplicação de metodologias inovadoras no estudo da zona costeira, como as Técnicas de Informação Geográfica (TIG), utilizando fotografia aérea e imagens de satélite de alta resolução espacial, é um assunto proeminente da investigação das áreas das Ciências Geo-Espaciais e da Engenharia Costeira. Um conjunto de fotografias aéreas, entre 1958 e 2002, foi analisado visualmente num ambiente de Sistemas de Informação Geográfica (SIG), com o objectivo de identificar hidroformas e hidromorfologias costeiras, no sector entre Esmoriz e Mira. Este trabalho tem como objectivo principal identificar e analisar formas/padrões morfológicas e hidrodinâmicos (hidroformas e hidromorfologias) recorrendo a algoritmos da classificação de imagem. Para alcançar esse objectivo foram aplicados diferentes métodos de classificação de imagem, nomeadamente técnicas de classificação supervisionada e não supervisionada, utilizando o software PCI Geomatica®. Foram testados diferentes algoritmos na classificação supervisionada, (paralelepípedo, distância mínima e máxima probabilidade) e na classificação não supervisionada, o K-médias e o ISODATA. Os algoritmos de classificação supervisionada apresentaram bons resultados, demonstrados pela precisão global e coeficiente Kappa, de 95.65% - 0.95661 e de 95.85% - 0.95840, para o método do paralelepípedo e para o método da máxima probabilidade respectivamente. Os algoritmos de classificação não supervisionada (K-médias e ISODATA) permitiram identificar várias classes, como por exemplo, praia, face da praia e zona de rebentação. Os resultados obtidos foram comparados (sobrepostos) com os da análise visual em ambiente SIG, mostrando uma boa concordância nas hidroformas e hidromorfologias identificadas.
Resumo:
Neste trabalho será apresentada a modelagem por regressão logística, com a finalidade de prever qual seria a inadimplência dos clientes que compõem o portfólio de uma grande instituição financeira do país. Sendo assim, será explorada a ideia de usar o conceito de provisionamento pura e simplesmente, através da estimação de uma probabilidade de default dado por um ou mais modelos estatísticos que serão construídos no decorrer do trabalho, conforme incentiva o comitê de Basileia. Um dos modelos será feito a partir de uma separação prévia de público através de clusters e a outra técnica a ser explorada será a criação de um modelo sem nenhuma separação. O objetivo será a comparação entre as duas métricas de classificação de risco e verificar os trade-off entre elas e os impactos de variáveis macroeconômicas nestes modelos.
Resumo:
Pós-graduação em Ginecologia, Obstetrícia e Mastologia - FMB
Resumo:
Knowing the annual climatic conditions is of great importance for appropriate planning in agriculture. However, the systems of climatic classification are not widely used in agricultural studies because of the wide range of scales in which they are used. A series with data from 20 years of observations from 45 climatological stations in all over the state of Pernambuco was used. The probability density function of the incomplete gamma distribution was used to evaluate the occurrence of dry, regular and rainy years. The monthly climatic water balance was estimated using the Thornthwaite and Mather method (1955), and based on those findings, the climatic classifications were performed using the Thornthwaite (1948) and Thornthwaite and Mather (1955) for each site. The method of Kriging interpolation was used for the spatialization of the results. The study classifications were very sensitive to the local reliefs, to the amount of rainfall, and to the temperatures of the regions resulting in a wide number of climatic types. The climatic classification system of Thornthwaite and Mather (1955) allowed efficient classification of climates and a clearer summary of the information provided. In so doing, it demonstrated its capability to determine agro climatic zones.
Resumo:
Inductively Coupled Plasma Optical Emission Spectrometry was used to determine Ca, Mg, Mn, Fe, Zn and Cu in samples of processed and natural coconut water. The sample preparation consisted in a filtration step followed by a dilution. The analysis was made employing optimized instrumental parameters and the results were evaluated using methods of Pattern Recognition. The data showed common concentration values for the analytes present in processed and natural samples. Principal Component Analysis (PCA) and Hierarchical Cluster Analysis (HCA) indicated that the samples of different kinds were statistically different when the concentrations of all the analytes were considered simultaneously.
Resumo:
Remotely sensed imagery has been widely used for land use/cover classification thanks to the periodic data acquisition and the widespread use of digital image processing systems offering a wide range of classification algorithms. The aim of this work was to evaluate some of the most commonly used supervised and unsupervised classification algorithms under different landscape patterns found in Rondônia, including (1) areas of mid-size farms, (2) fish-bone settlements and (3) a gradient of forest and Cerrado (Brazilian savannah). Comparison with a reference map based on the kappa statistics resulted in good to superior indicators (best results - K-means: k=0.68; k=0.77; k=0.64 and MaxVer: k=0.71; k=0.89; k=0.70 respectively for three areas mentioned). Results show that choosing a specific algorithm requires to take into account both its capacity to discriminate among various spectral signatures under different landscape patterns as well as a cost/benefit analysis considering the different steps performed by the operator performing a land cover/use map. it is suggested that a more systematic assessment of several options of implementation of a specific project is needed prior to beginning a land use/cover mapping job.
Resumo:
The creation of the Brazilian Program for the Modernization of the Horticulture by the Secretariat of Agriculture and Supplying of the State of São Paulo at CEAGESP, determined the standardization of fruit and vegetables in the follow aspects: degree of coloration, format, calibers, defects and packing. Therefore, the main goal of this research is to correlate the classification given by the Brazilian Program with the one used by the wholesalers at CEAGESP, verifying if the established norms are being fulfilled for cultivar Carmen and Debora (SAKATA SEED). The results showed, that for cultivar Carmem, for the averages of the observed values it does not move away from the norms created by the Program for sizes small and medium. However, for the case of cultivar Debora, the results showed differences between the adopted classifications. The tomatoes were devaluated, because had been commercialized below of the standardization indicated for the Brazilian Program.
Resumo:
Quality evaluation of classification was done in two fresh market tomatoes packing house, using electronically and mechanical equipments in two harvest periods, summer and winter seasons. The main goal of this work was to evaluate size and color grading conformity with the standards proposed by the Brazilian Program for Horticulture Modernization and size grading obtainded with the one established by the packer. The cultivar studied was Carmen. The results showed that there was no grade conformity with the fresh tomato quality standards proposed by the Brazilian Program for Horticulture Modernization. The grade conformity obtained when compared with the one programmed by the packer, was only for large sizes, in both equipments. The electronically equipment has presented better performance, over the mechanical, considering grading quality and fruits post-harvest quality. However, the electronically equipment must be constantly monitored to achieve efficiency and investment return. On the other side, for mechanical equipment it will be necessary to review the actual system of size grading, in order to follow the fresh tomato quality standards.
Resumo:
On the last years, in Brazil, sorting and classifying fruits and vegetables using packing lines have increased. This work aimed at characterizing the cleaning process for fresh market tomatoes at two packing lines, one imported and one national located at Campinas, São Paulo State. Characterization included data, number, types and brushes velocity, water use, fruit standing time and cleaning efficiency. Standing time was measured correlating to fruit diameter (CEAGESP). For measuring cleaning efficiency an equipment was developed that was mainly composed of a ring involved with white cloth. Samples were taken before and after the cleaning step and evaluated using a colorimeter HUNTER Lab. The results showed a strong difference between the two equipments. The imported equipment showed lower number on brushes and rotation than national one, however a higher water consumption. For imported equipments this relation was not found. Both packing lines showed the same cleaning efficiency. Cleaning efficiency is related to be an interaction among the studies parameters, and it could be necessary a better management than the one used on both equipments.
Resumo:
Universidade Estadual de Campinas. Faculdade de Educação Física