Thesis (Ph. D.)--University of Texas at El Paso, 2008. / Title from title screen. Vita. CD-ROM. Includes bibliographical references. Also available online.

Thesis (Ph. D.)--University of Missouri-Columbia, 1999. / Typescript. Vita. Includes bibliographical references (leaves 36-44). Also available on the Internet.

Thesis (M.S.)--State University of New York at Binghamton, Thomas J. Watson School of Engineering and Applied Science, Department of Computer Science, 2009. / Includes bibliographical references.

Σε αυτή την εργασία ασχοληθήκαμε με δύο μεθόδους, που σκοπός τους είναι να κατατάσσουν τις παρατηρήσεις σε γνωστές ομάδες και στη συνέχεια να κάνουν προβλέψεις για καινούριες παρατηρήσεις. Αυτές οι μέθοδοι είναι η λογιστική παλινδρόμηση (logistic regression) και η διαχωριστική ανάλυση (discriminant analysis). Στο πρώτο κεφάλαιο αναφέραμε περιληπτικά τα μη γραμμικά μοντέλα παλινδρόμησης (αφού και η λογιστική παλινδρόμηση είναι ένα τέτοιο μοντέλο). Απλά αναφέρουμε τη μορφή που έχουν αυτά τα μοντέλα, με ποιες μεθόδους μπορούμε να εκτιμήσουμε τις παραμέτρους παλινδρόμησης, ποια είναι τα διαστήματα εμπιστοσύνης για τους συντελεστές παλινδρόμησης και τη μορφή που θα έχουν οι έλεγχοι υποθέσεων. Στο δεύτερο κεφάλαιο περιγράφουμε τη λογιστική παλινδρόμηση. Η λογιστική παλινδρόμηση είναι χρήσιμη σε καταστάσεις στις οποίες επιθυμούμε να προβλέψουμε την ύπαρξη ή την απουσία ενός χαρακτηριστικού ή ενός συμβάντος. Η πρόβλεψη αυτή βασίζεται στην κατασκευή ενός μοντέλου και συγκεκριμένα στον προσδιορισμό των τιμών που παίρνουν οι συντελεστές. Αυτή η μέθοδος είναι μια γενίκευση της απλή γραμμικής παλινδρόμησης για την περίπτωση όπου η εξαρτημένη μεταβλητή είναι δίτιμη (παίρνει την τιμή 0 όταν το χαρακτηριστικό απουσιάζει και την τιμή 1 όταν υπάρχει το χαρακτηριστικό). Στο τρίτο κεφάλαιο αναλύουμε τη διαχωριστική ανάλυση, η οποία έχει δύο στόχους: να χωρίσει ένα πληθυσμό σε ευδιάκριτες ομάδες και με τη βοήθεια ενός διαχωριστικού κανόνα να κατατάσσει παρατηρήσεις στις ευδιάκριτες ομάδες. Στο τέλος του κεφαλαίου περιγράφουμε τις ομοιότητες και τις διαφορές της διαχωριστικής ανάλυσης και της λογιστικής παλινδρόμησης. Στο τέταρτο και τελευταίο κεφάλαιο απλά δίνουμε ένα παράδειγμα που το λύνουμε με τη μέθοδο της λογιστικής παλινδρόμησης και ένα παράδειγμα που το λύνουμε με τη μέθοδο της διαχωριστικής ανάλυσης. Αυτό το κάνουμε με τη βοήθεια του στατιστικού πακέτου SPSS. / In this work we dealt with two methods, that their aim are to classify the observations in known teams and afterwards to make forecasts for new observations. These methods are the accountant regression (logistic regression) and the bisector analysis (discriminant analysis).

Made available in DSpace on 2014-06-11T19:26:55Z (GMT). No. of bitstreams: 0 Previous issue date: 2009-04-30Bitstream added on 2014-06-13T19:55:22Z : No. of bitstreams: 1 nunes_rso_me_sjrp.pdf: 433378 bytes, checksum: 58b9cb2328012880486c9f1011d4714a (MD5) Bitstreams deleted on 2014-12-19T18:32:50Z: nunes_rso_me_sjrp.pdf,Bitstream added on 2014-12-19T18:33:36Z : No. of bitstreams: 1 000592541.pdf: 548568 bytes, checksum: df983105ba2c1be9b281439d3ea72a48 (MD5) / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Dado um número inteiro positivo d, encontrar um corpo de grau d que tenha, em valor absoluto, o menor discriminante é um problema clássico e poucos resultados se tem até hoje no sentido de se resolver tal desafio. O principal interesse deste trabalho consiste em estudar o problema acima sobre os corpos de números Abelianos, particularmente aqueles de grau primo. Para tanto será preciso dominar algumas técnicas referentes ao cálculo do discriminante de corpos de números, em especial, dos corpos Abelianos. / Given a positive interger d, finding a field of degree d which has, to absolute value, the smallest discriminant it is a classical problem and few results has been got until at present time, to solve this challenge. The main purpose of this paper it is to study the problem above on Abelian numbers fields in special that ones of prime degree. However, it is necessary to know any techniques for calculating the numbers fiel discriminant, specially, to Abelian fields.

Made available in DSpace on 2014-06-11T19:26:56Z (GMT). No. of bitstreams: 0 Previous issue date: 2006-02-17Bitstream added on 2014-06-13T20:47:38Z : No. of bitstreams: 1 quilles_cro_me_sjrp.pdf: 739363 bytes, checksum: b87b737abd0925a66989e0972946f214 (MD5) / O objetivo deste trabalho e mostrar duas maneiras de se calcular o discriminante de um corpo de números. Da primeira forma, utilizando a teoria algébrica dos números clássica vimos como calcular o discriminante dos corpos quadráticos e corpos ciclotômicos. Através desta teoria é possível calcular o discriminante somente desses corpos com um árduo trabalho. Da segunda maneira utilizando os caracteres de Dirichlet e seus condutores vimos o cálculo do discriminante para qualquer corpo abeliano de uma maneira não muito trabalhosa. Finalmente, utilizando esses resultados damos aplicações sobre reticulados algébricos. / The aim of this work is to make a parallel between two forms of computing discriminants of fields of numbers. In the first form, by classic algebraic number theory we computed the discriminant of quadratics fields and ciclotomic fields. Through of this theory, is possible to computing the discriminant alone of this fields with a arduous work. In the second form using Dirichlet's character and their conductors we computed the discriminant of any abelian field of a form not very hard. Finally, using this results we give applications on algebraic lattices.

Made available in DSpace on 2015-03-26T13:32:07Z (GMT). No. of bitstreams: 1 texto completo.pdf: 1250297 bytes, checksum: 6fd0be6a8317d32468f8f7b7ff74d7ef (MD5) Previous issue date: 2009-02-17 / Fundação de Amparo a Pesquisa do Estado de Minas Gerais / The concern in minimizing the amount of chemical products used in farmings is increasing. The use of artificial vision systems has been demonstrating a great potential for use of varied taxes of inputs, as for instance, the application herbicides only in places where the presence of harmful plant is detected. The good acting of a system developed for this purpose depends mainly of the characteristic use that they allow to differentiate patterns of harmful plants of the pattern of the cultivated species. Like this, the objective of the present work was to develop and to evaluate a characteristic for the recognition of the patterns corn plant and harmful plant. The specific objectives were: the) to identify which image, green excess or the index of vegetation of normalized green, tends to provide better classification; b) to compare the classification obtained by characteristics geoestatistics, obtained when using the characteristics of Haralick. With this purpose, were acquired to the 29 days after the emergency, period in that it is usually made the application of herbicides, nine corn images (Zea Mays L.) and of each one of the species of appraised harmful plants in this experiment: Euphorbia heterophylla L., Digitaria horizontalis Willd, Cenchrus chinatus L. Six of these images were used for the selection of the characteristic that promotes better acting in the classification. The remaining three were used for the validation of the selected characteristic. Each one of the six training images was cut out in 100 blocks of 68x68 pixels. For each one of the blocks was obtained the value of the characteristic textural geoestatistic (variogram, the madogram, cross variogram and pseudo cross variogram) and the one of Haralick (angular moment, average, variance, entropy, correlation, moment of the product, inverse moment of the difference and correlation measures). Additionally, characteristic geoestatísticos and no-geoestatísticos they were obtained considering different angles (0, 45, 90 and 135°) of relationship among pixels. Characteristic geoestatistics were, also, obtained for different distances (1, 2, 3, 4, 5, 6, 7, 8, 9, 10) of in pairs among pixels. The characteristic variograma and madograma were calculated to leave of the image green excess and GNDVI. Already the characteristic cross variogram and pseudo cross variogram were calculated with Greenness Method use in the blocks using the combinations of the bands RxG, GxB and IVxG. The characteristic of Haralick were calculated starting from the images of the green excess and GNDVI. The acting of the characteristic, proposed like this, it was evaluated using discriminate analysis. The selected characteristic were those that presented larger value for the index kappa. Additionally, new characteristic were obtained starting from combinations of the selected characteristic. These combinations, also, had appraised acting using the discriminant analysis with the objective of to identify which combination provides better classification. Later, the power of generalization of the selected combination was evaluated using the three images of each species reserved for the validation stage. The conclusions obtained regarding the objectives proposed in this research were a) the image that tended to present the best results of the index kappa was the image excess of green; b) the characteristic obtained starting from the function madograma and the one of Haralick were the ones that supplied the best results; c) the characteristic geoestatistic madograma in the 10 distances and angle 0° presented better classification results when used without combination of other characteristic; d) the characteristic geoestatísticos and the one of Haralick, when used separately didn't present such good as combined results; e) the characteristic use that consider the continuity of the pixel values, in the recognition of patterns can be a fundamental tool in the classification process. / A preocupação em minimizar a quantidade de produtos químicos utilizado em lavouras vem aumentando. O uso de sistemas de visão artificial tem demonstrado um grande potencial para o uso de taxas variadas de insumos, como por exemplo, a aplicação de herbicidas somente em locais onde é detectada a presença de planta daninha. O bom desempenho de um sistema desenvolvido para esta finalidade depende principalmente do uso de descritores que permitam diferenciar padrões de plantas daninhas do padrão da espécie cultivada. Sendo assim, objetivo geral do presente trabalho foi desenvolver e avaliar um descritor para o reconhecimento dos padrões planta de milho e planta daninha. Os objetivos específicos foram: a) identificar qual imagem, excesso de verde ou o índice de vegetação de verde normalizado, tende a proporcionar melhor classificação; b) comparar a classificação obtida por descritores geoestatísticos, com a obtida ao usar os descritores de Haralick. Com esta finalidade, foram adquiridas aos 29 dias após a emergência, período em que normalmente é feita a aplicação de herbicidas, nove imagens de milho (Zea Mays L.) e de três espécies de plantas daninhas avaliadas neste experimento: leiteira (Euphorbia heterophylla L.), capim-milhã (Digitaria horizontalis Willd) timbête (Cenchrus echinatus L.). Seis destas imagens foram utilizadas para a seleção do descritor que promove melhor desempenho na classificação. As três restantes foram utilizadas para a validação do descritor selecionado. Cada uma das seis imagens de treinamento foi recortada em 100 blocos de 68x68 pixels. Para cada um dos blocos foi obtido o valor dos descritores texturais geoestatísticos (variograma, o madograma, variograma cruzado e pseudo variograma cruzado) e os de Haralick (momento angular, média, variância, entropia, correlação, momento do produto, momento inverso da diferença e medidas de correlação). Adicionalmente, descritores geoestatísticos e não-geoestatísticos foram obtidos considerando diferentes ângulos (0, 45, 90 e 135°) de relacionamento entre pixels. Descritores geoestatísticos foram, também, obtidos para diferentes distâncias (1, 2, 3, 4, 5, 6, 7, 8, 9, 10) de pareamento entre pixels. Os descritores variograma e madograma foram calculados partir da imagem excesso de verde e GNDVI. Já os descritores variograma cruzado e pseudo variograma cruzado foram calculados com o uso do Greenness Method nos blocos usando as combinações das bandas RxG, GxB e IVxG. Os descritores de Haralick foram calculados a partir das imagens do excesso de verde e GNDVI. O desempenho dos descritores, assim propostos, foi avaliado usando análise discriminante. Os descritores selecionados foram aqueles que apresentaram maior valor para o índice kappa. Adicionalmente, novos descritores foram obtidos a partir de combinações dos descritores selecionados. Estas combinações, também, tiveram o seu desempenho avaliado usando a análise discriminante com o objetivo de identificar qual combinação proporciona melhor desempenho na classificação. Posteriormente, o poder de generalização da combinação selecionada foi avaliado usando as três imagens de cada espécie reservadas para a etapa de validação. As conclusões obtidas com relação aos objetivos propostos nesta pesquisa foram a) a imagem que tendeu a apresentar os melhores resultados do índice kappa foi a imagem excesso de verde; b) os descritores obtidos a partir da função madograma e os de Haralick foram os que forneceram os melhores resultados; c) o descritor geoestatístico madograma nas 10 distâncias e ângulo 0° apresentou melhores resultados de classificação quando usado sem combinação de outros descritores; d) os descritores geoestatísticos e os de Haralick, quando usados isoladamente não apresentaram resultados tão bons quanto combinados; e) o uso de descritores que consideram a continuidade dos valores de pixel, no reconhecimento de padrões pode ser uma ferramenta fundamental no processo de classificação.

Made available in DSpace on 2015-03-26T13:32:08Z (GMT). No. of bitstreams: 1 texto completo.pdf: 958045 bytes, checksum: a7576f2d825f5b6d63c2110fc9afce32 (MD5) Previous issue date: 2009-12-15 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / The correct classification of individuals in pre-established groups has become of great importance in breeding. The multivariate statistical techniques commonly used in this type of problem are the discriminant functions of Fisher and Anderson, which are used to allocate an initially unknown individual in one of the g populations or pre-defined groups. In recent decades a new computing paradigm, artificial neural networks, has come along to solve various problems of Statistics, such as grouping of similar individuals, time series forecasting and also of particular interest, the problem of classification. The objective of this research was to conduct a simulation study in order to compare the discriminant functions of Fisher and Anderson and neural networks. We evaluated the number of incorrect classifications of individuals known to belong to different populations with different levels of dissimilarity measured by the Mahalanobis distance. Simulations were conducted using the software Genes (Cruz, 2006). Although Artificial Neural Networks presented a rate of incorrect classification of individuals rejected for being considered ambiguous as to its discriminatory characteristics, it proved to be a promising technique, since it presented a lower number of incorrect classifications of individuals when compared to the discriminant functions. / A correta classificação de indivíduos em grupos pré-estabelecidos tem se tornado de grande importância no melhoramento genético. As técnicas de estatística multivariada usualmente utilizadas nesse tipo de problema são as funções discriminantes de Fisher e as funções discriminantes de Anderson, que são usadas para alocar um indivíduo inicialmente desconhecido em uma das g populações ou grupos pré-definidos. Nas últimas décadas vêm surgindo um novo paradigma de computação, as redes neurais artificiais, que podem ser utilizadas para resolver diversos problemas da Estatística, como agrupamento de indivíduos similares, previsão de séries temporais e em especial, os problemas de classificação. O objetivo dessa pesquisa foi realizar um estudo comparativo entre as funções discriminantes de Fisher e de Anderson e as redes neurais artificiais quanto ao número de classificações erradas de indivíduos sabidamente pertencentes a diferentes populações, com distintos níveis de dissimilaridade. Essa dissimilaridade, medida pela distância de Mahalanobis, foi um conceito de fundamental importância na utilização das técnicas de discriminação, pois quantificou o quanto as populações eram divergentes. Quanto maior o valor observado para essa medida, menos similares foram as populações em análise. A obtenção dos dados foi feita através de simulação utilizando o programa computacional Genes (CRUZ, 2006). As redes neurais artificiais apresentaram uma taxa de indivíduos rejeitados por serem considerados ambíguos quanto às suas características discriminatórias. No entanto, mostraram-se uma técnica promissora no que diz respeito a problemas de classificação, uma vez que apresentaram um número de classificações erradas de indivíduos menor que aqueles dados pelas funções discriminantes.

Sex estimation in bioarchaeological research is paramount for comparative analyses of skeletal remains and developing estimates of other demographic variables. Frequently, sex is estimated utilizing the morphology of pelvic and cranial bones in archaeological skeletal samples. Regrettably, these bones are often damaged, destroyed, or lost and cannot be employed for the estimation of sex. Fortunately, a variety of metric and visual sex estimation methods have been developed on modern skeletal samples with known demography. Disappointingly, due to the population specific nature of many of these metric methods, they cannot be accurately applied directly to alternate skeletal samples. However, these methods can be redeveloped and retested for archaeological skeletal samples if the proper protocol is utilized. This research utilizes the protocol for developing methods of sex estimation on samples with unknown sex developed by Murail et al. (1999). Utilizing select hand and foot bones from four prehistoric Native American skeletal samples from Alabama and Illinois, multiple discriminant functions were developed and tested on both Archaic and Mississippian Period skeletal samples. Furthermore, the four individual site samples were combined into two geographical and two temporal samples, as well as an all pooled sample in order to test the broader applicability of these methods. The results indicate that sex estimation methods can be developed on samples where sex is not known. Additionally, the discriminant functions developed produce high levels of classification for the sites, individually, as well as for the geographic, temporal, and all pooled samples. The latter suggests these functions have broader applicability for these regions and temporal periods. Moreover, the concordance rates for these functions are similar to the accuracy rates when these bones were applied to samples with known sex.

This project analyzes the sexual dimorphism of the femur and tibia using geometric morphometrics. The study sample includes 250 individuals of known sex and age at death with complete, non-damaged, non-pathological skeletal remains from the William M. Bass Donated Skeletal Collection at the University of Tennessee, Knoxville. Ages range from 19-96 for males (mean=56.92 years) and 29-97 for females (mean=59.48 years). A combination of landmarks and semi-landmarks were collected on the proximal and distal epiphyses of each bone using a Microscribe, which helps capture the overall size and shape variation present in the sample. Only individuals from one population, White, where analyzed in order to eliminate population variation bias. Classification rates for males and females for the proximal femur were 80.8% and 78.4% respectively, for the distal femur 92.6% and 89.6% respectively, for the proximal tibia 80.8% and 83.2% respectively, and the distal tibia 81.6% and 80.8% respectively, all with a p<0.0001. These rates created a classification model for which epiphysis gave the most accurate assessment of sex: the distal femur, followed by the proximal tibia, then the distal tibia, and lastly the proximal femur. This study indicates the knee joint is the most dimorphic, followed by the ankle and then the hip. The results fall in line with another study indicating the knee is more sexually dimorphic in a modern White population (Spradley and Jantz 2011), though in contrast to their results this study found the distal femur was more dimorphic than the proximal tibia. This method indicates that in comparison to standard measurements, geometric morphometrics may provide a more reliable method for sex estimation when used, specifically on the knee. Certain landmarks were then selected based on the standard taphonomic process of coffin wear and postmortem damage (Pokines and Baker 2014) for exclusion to determine the usability of the method on fragmented or damaged skeletal remains. When combinations of landmarks were removed, the distal femur still possessed the highest classification rates with over 80% accuracy.