This thesis critically analyzes some measures of masculinity and femininity. Special attention is paid primarily to a BSRI (Bem Sex-Role Inventory), which is the most widespread but completely invalid instrument for measuring gender identity. At the same time, there is presented an alternative approach to the topic of gender identity based on an application of statistical discriminant analysis technique to generally shared stereotypical evaluations of typical men and women. In empirical part, proposed approach is successfully tested in practice by creating an index of masculinity / femininity. Results of analyzes of relations between the determined gender identity and various other components of gender belief system (cognitive gender schematization, an explicit M / F and gender attitudes) suggest the validity of this approach. Between feminity (regardless of sex) and a measure of traditional gender attitudes has been found small but significant negative correlation (rs = -0.3). Keywords: masculinity, feminity, discriminant analysis, critique of BSRI

Doenças como AVC e Epilepsia estão constantemente associadas à manifestação de comprometimentos cognitivos e demência, tendo como característica principal os déficits de memória. Atualmente,o instrumento Mini Exame do Estado Mental (MEEM) é o teste de rastreio de comprometimento cognitivo mais utilizado no mundo, individualmente ou como parte integrante em baterias de avaliação neuropsicológica. Devido afalta de instrumentos específicos pararastreiode comprometimento cognitivo de indivíduos com sintomas de demênciavítimas de AVC e/ou Epilepsia em Angola, África, fez se necessária aadaptação evalidação de instrumentos confiáveis.O presente estudoteve por objetivo a avaliação clínicada versão brasileira do instrumento MEEM adaptado por Brucki, para uso na população adulta de Angola.Trata-se de um estudo clínico comparado, envolvendo três populações. Foram incluídos neste estudo 85 indivíduos acometidos por AVC, 40 por Epilepsia e como grupo de comparação, 87 indivíduos de 20 a 65 anos de idade, supostamente saudáveis. A escolaridade mínima para inclusão do estudo foi definida em 4 anos de ensino formal. O presente estudo foi desenvolvido em parceria com o Serviço de Neurologia do Hospital Américo Boavida e com a Faculdade de Medicina da Universidade Agostinho Neto (FMUAN), ambos localizados ao norte da cidade de Luanda, capital de Angola, África.Toda a amostra foi submetida à coleta de dados sócio demográficos e avaliação neuropsicológica, através do instrumento MEEM. O instrumento consistiu no rastreio de comprometimento cognitivo, contemplando funções como orientação temporal e espacial, memória, cálculo, linguagem e capacidade construtivavisual.No presente estudo, encontramos dados que apontam um desempenho significativamente pior das funções cognitivas em pacientes acometidos por AVC quando comparados a pacientes com Epilepsia e grupo dossaudáveis. Esses achados confirmam a validade do instrumento MEEM, em função das características apresentadas pela doença, sendo possível discriminar com clareza os grupos AVC, Epilepsia e Controles Saudáveis. Os dados obtidos indicam uma boa qualidade da versão na língua portuguesa angolana, sendo a consistência interna da versão adaptada com ? de Cronbach = 0,736, semelhante aos dados de referência. Este estudo pioneiro preenche uma importante lacuna na produção de conhecimento em Angola, abrindo um leque de possibilidades para outros trabalhos científicos no país, facilitando a aplicação dos resultados tanto na área acadêmica-científica, quanto na clínica. É necessário aos profissionais de saúde que se apropriem de mais conhecimentos, para que possam propor intervenções adequadas, que auxiliam nos direitos de saúde e bem-estarda população angolana. / Diseases such as stroke and epilepsy are constantly associated with the manifestation of cognitive impairment and dementia, having as main feature the memory deficits (Caramelli & Castro, 2005). Currently, the instrument Mini Mental State Examination (MMSE) is the screening test of cognitive impairment most used in the world, individually or as an integrated part in neuropsychological assessment battery (Melo & Barbosa, 2015). Due to the lack of specific instruments for screening cognitive impairment of individuals with dementia symptoms victims of stroke and / or epilepsy in Angola, Africa, it was necessary to adapt and validate reliable instruments. The presentestudy had as objective the clinical evaluationof the brazilian version of the MMSE instrument adapted by Brucki et al. (2003), for use in the adult population of Angola, Africa. It is a comparative clinical study involving three populations. This studyincluded 85 individuals affected by stroke, 40 by Epilepsy, and as a comparison group, 87 individuals, 20 to 65 years of age, who were supposed to be healthy. The minimum schooling for inclusion of the study was defined in 4 years of formal education. Thepresent study was developed in partnership with the Neurology Service of the Américo Boavida Hospital and the Faculty of Medicine of the Agostinho Neto University (FMUAN), both located north of the city of Luanda, the capital of Angola, Africa. All the sample was submitted to the collection of sociodemographic data and neuropsychological evaluation, through the MMSE instrument. The instrument consisted in the screening of cognitive impairment, including functions such as temporal and spatial orientation, memory, calculus, language and visual constructive capacity.In the presente study, we found data that indicate a significantly worse performance of cognitive functions in patients with stroke when compared to patients with epilepsy and healthy group. These findings confirm the validity of theinstrument MMSE in, due to the characteristics of the disease, being possible to discriminate with clarity the groups Stroke, Epilepsy and Healthy Controls. The data obtained indicate a good quality of the Angolan Portuguese language version, being the internal consistency of the adapted versionwhit the Cronbach\'s alpha = 0.736, similar to the reference data.This pioneering study fills animportant gap in knowledge production in Angola, opening up a range of possibilities for other scientific papers in the country, facilitating the application of the results both in the academic and scientific area, as at the clinic. It is necessary that health professional take ownership of more knowledge, so that they can propose appropriate interventions that assist in health and welfare rights of the Angolan population.

This paper presents a description of gray-headed junco habitat in the form of a model based on discriminant function analysis. Junco nests were found by searching randomly located grids on a 7.8 km2 study area in central Utah. Vegetation data was gathered on 500 m2 circular plots surrounding nests and contrasted with similar data from searched grids where nests were not found. The model explains 28 percent of the between-groups variance and correctly classifies 68 percent of the plots. Plant cover types are good predictors of areas where gray-headed juncos nest, while plant community type descriptions do not distinguish between utilized and unutilized areas. The use of quantitative models of wildlife habitat with data derived from various kinds of information systems is discussed.

Mitochondrial DNA (mtDNA) has played a major role in human population studies over the past decade due to its maternal inheritance and negligible recombination (Macaulay, 1999). The mtDNA control region has been the focus of these studies due to the highly polymorphic nature of this non-coding region. Forensic scientists also use mtDNA to help determine the identity of missing individuals when nuclear DNA is not present. However, when skeletal remains are unclaimed, identification becomes near impossible. Therefore, mtDNA can play a valuable role in identification in terms of population affiliation, especially in conjunction with morphological analysis. The goals of this research were two-fold: 1) to determine population affiliation of unknown skeletal samples using phylogenetics and 2) to find a method of extraction that leaves a majority of the remains intact. This research depended on the donation of samples from sixteen skeletal remains from the Hillsborough County Medical Examiners office. Mitochondrial DNA from ten of these cases were extracted, amplified, and sequenced in order to determine population affiliation via phylogenetic analysis of hypervariable region I (HVR I). These sequences were aligned and compared to that of sequences in a pre-existing mtDNA control region database (Handt, 1998). The crania of the skeletal remains were measured and subsequently analyzed by the forensic anthropology program FORDISC 2.0 to morphologically determine population affiliation. A secondary morphological analysis included input of the measurements into SPSS, a statistical program package, as a separate discriminant function assessment. This analysis was dependent on a database of craniometrics from known individuals (Jantz, and Moore-Jansen, 2000). The results were compared to those of FORDISC 2.0 as well as that of the phylogenetic tree constructed using molecular techniques. This study will determine whether phylogenetic analysis is a legitimate way to determine population affiliation of unknown individuals, thereby benefiting future forensic studies.

Behavioral characterization of animal models for Alzheimer's Disease is critical for the development of potential therapeutics and treatments against the disease. While there are several known animal models of AD, three current models--APPsw, P301L, and APPsw+P301L--have not been well characterized, if at all. This study, therefore, aimed to perform a full behavioral characterization of these three models in order to better understand the impairments associated with each one. Between 5 and 8.5 months of age, animals were behaviorally tested in a variety of sensorimotor, anxiety, and cognitive tasks. The number of tau+ neurons in the forebrains of P301L mice was then compared to their behavioral performance. Results of this study indicate that retinal degeneration (rd) seriously impairs the performance of mice in behavioral tasks. Animals that carry the homozygous allele of this mutation must, therefore, be eliminated from any such study requiring visual acuity. After this elimination, my findings indicate that APP mice are impaired in several cognitive tasks (including platform recognition, Morris maze, Y-maze, and radial-arm water maze) at a young early age (5 to 8.5 months of age). These mice have fairly normal sensorimotor function, showing significant impairment only in balance beam performance starting at 5 months. Although P301L mutant Tau mice, as a group, did not have significant impairments in any sensorimotor or cognitive task, correlation analysis revealed that higher numbers of tau+ neurons in cortex and hippocampus were associated with poorer cognitive performance. Finally, discriminant function analysis (DFA) appears able to accurately discriminate between the three transgenic groups of mice using only an 8-measure data set. In conclusion, this study provides the first comprehensive, multiple task behavioral assessment of the APPsw and P301L animal models of AD, indicating that APPsw mice are cognitively impaired at an early age while P301L mice are largely unimpaired through 8.5 months. Nonetheless, correlational analysis implicates the formation of neurofibrillary tangles in the onset of cognitive impairments. Finally, my findings recommend the continued use of DFA to determine if groups of animals, based on different transgenicity or therapeutic treatment, could be discriminated between from their behavior alone.

In many pattern recognition applications, a large number of features are extracted in order to ensure an accurate classification of unknown classes. One way to solve the problems of high dimensions is to first reduce the dimensionality of the data to a manageable size, keeping as much of the original information as possible and then feed the reduced-dimensional data into a pattern recognition system. In this situation, dimensionality reduction process becomes the pre-processing stage of the pattern recognition system. In addition to this, probablility density estimation, with fewer variables is a simpler approach for dimensionality reduction. Dimensionality reduction is useful in speech recognition, data compression, visualization and exploratory data analysis. Some of the techniques which can be used for dimensionality reduction are; Factor Analysis (FA), Principal Component Analysis(PCA), and Linear Discriminant Analysis(LDA). Factor Analysis can be considered as an extension of Principal Component Analysis. The EM (expectation maximization) algorithm is ideally suited to problems of this sort, in that it produces maximum-likelihood (ML) estimates of parameters when there is a many-to-one mapping from an underlying distribution to the distribution governing the observation, conditioned upon the obervations. The maximization step then provides a new estimate of the parameters. This research work compares the techniques; Factor Analysis (Expectation-Maximization algorithm based), Principal Component Analysis and Linear Discriminant Analysis for dimensionality reduction and investigates Local Factor Analysis (EM algorithm based) and Local Principal Component Analysis using Vector Quantization.

La présente étude vise à vérifier la conjecture faite par B. Gross relative à l'existence de corps de nombres de groupe de Galois non résoluble et ramifiés en un unique premier p < 11.<br />À travers ce travail, nous nous intéressons au cas des corps de nombres de degré n ≤ 9. Après quelques rappels généraux sur les outils utilisés, on présente les méthodes pratiques permettant de vérifier cette conjecture.<br />Les travaux de J. Jones ont montré que les corps de nombres de degré 5 et 6 vérifiant ces types de ramification ont tous un groupe de Galois résoluble.<br />Dans le cas du degré 7, S. Brueggeman a abouti au même résultat que le travail sus cité.<br />Nos travaux dans le cas des degrés 8 et 9 montrent que sous GRH ou de façon inconditionnelle, la ramification en 5 n'est pas possible. À l'issue des recherches numériques, les seules tables obtenues sont celles de la ramification en p = 2 en degré 8 et celles de la ramification en p = 3 en degré 9. Les corps obtenus ont tous un groupe de Galois résoluble, montrant ainsi que cette conjecture de B. Gross n'est pas vérifiée pour les corps de nombres de degré n ≤ 9.

AdaBoost.M1 has been successfully applied to improve the accuracy of a learning algorithm for multi-class classification problems. However, it assumes that the performance of each base classifier must be better than 1/2, and this may be hard to achieve in practice for a multi-class problem. A new algorithm called AdaBoost.MK only requiring base classifiers better than a random guessing (1/k) is thus designed. Early SVM-based multi-class classification algorithms work by splitting the original problem into a set of two-class sub-problems. The time and space required by these algorithms are very demanding. In order to have low time and space complexities, we develop a base classifier that integrates one-class SVMs with discriminant functions. In this study, a hybrid method that integrates AdaBoost.MK and one-class SVMs with improved discriminant functions as the base classifiers is proposed to solve a multi-class classification problem. Experimental results on data sets from UCI and Statlog show that the proposed approach outperforms many popular multi-class algorithms including support vector clustering and AdaBoost.M1 with one-class SVMs as the base classifiers.

When the huge building stock in Turkey is considered, it is practically impossible to carry out detailed structural analyses for all of the buildings. In order to cope with the seismic safety evaluation of a large number of existing buildings, it is necessary to use simplified techniques, which can predict the seismic vulnerability of the existing buildings in a relatively short time. The comprehensive structural data compiled for the 48 different reinforced concrete buildings contain full information on their structural characteristics before and after retrofitting and are used in this study. The first basic goal of the study is to develop a procedure through which the building stock under consideration can be classified as &ldquo / safe&rdquo / or &ldquo / unsafe&rdquo / according to the current Turkish Seismic Code. The classification procedure is based on discriminant analysis. The cross-sectional area of the load-bearing members of a building and its preliminary assessment score are selected as the discriminator variables. The second and ultimate basic goal of the study is to propose a method through which the minimum retrofitting cost for satisfying the provisions of the Turkish Seismic Code can be estimated. A quick and uncostly assessment of retrofitting cost estimates based on the procedure described in this thesis will provide a useful input for decisions concerning whether a seismically &ldquo / unsafe&rdquo / building should be rebuilt or retrofitted. Such a situation will save time, labor and money, when it is used for the evaluation of building stocks involving large number of buildings and also in urban transformation operations.

In this dissertation we study the feature selection and classification problems and apply our methods to real-world medical and biological data sets for disease diagnosis. Classification is an important problem in disease diagnosis to distinguish patients from normal population. DAMIP (discriminant analysis -- mixed integer program) was shown to be a good classification model, which can directly handle multigroup problems, enforce misclassification limits, and provide reserved judgement region. However, DAMIP is NP-hard and presents computational challenges. Feature selection is important in classification to improve the prediction performance, prevent over-fitting, or facilitate data understanding. However, this combinatorial problem becomes intractable when the number of features is large. In this dissertation, we propose a modified particle swarm optimization (PSO), a heuristic method, to solve the feature selection problem, and we study its parameter selection in our applications. We derive theories and exact algorithms to solve the two-group DAMIP in polynomial time. We also propose a heuristic algorithm to solve the multigroup DAMIP. Computational studies on simulated data and data from UCI machine learning repository show that the proposed algorithm performs very well. The polynomial solution time of the heuristic method allows us to solve DAMIP repeatedly within the feature selection procedure. We apply the PSO/DAMIP classification framework to several real-life medical and biological prediction problems. (1) Alzheimer's disease: We use data from several neuropsychological tests to discriminate subjects of Alzheimer's disease, subjects of mild cognitive impairment, and control groups. (2) Cardiovascular disease: We use traditional risk factors and novel oxidative stress biomarkers to predict subjects who are at high or low risk of cardiovascular disease, in which the risk is measured by the thickness of the carotid intima-media or/and the flow-mediated vasodilation. (3) Sulfur amino acid (SAA) intake: We use 1H NMR spectral data of human plasma to classify plasma samples obtained with low SAA intake or high SAA intake. This shows that our method helps for metabolomics study. (4) CpG islands for lung cancer: We identify a large number of sequence patterns (in the order of millions), search candidate patterns from DNA sequences in CpG islands, and look for patterns which can discriminate methylation-prone and methylation-resistant (or in addition, methylation-sporadic) sequences, which relate to early lung cancer prediction.