The study of facial expressions in humans dates back to the 19th century and the study of the emotions that these facial expressions portray dates back even further. It is a natural part of non-verbal communication for humans to pass across messages using facial expressions either consciously or subconsciously, it is also routine for other humans to recognize these facial expressions and understand or deduce the underlying emotions which they represent. Over two decades ago and following technological advances, particularly in the area of image processing, research began into the use of machines for the recognition of facial expressions from images with the aim of inferring the corresponding emotion. Given a previously unknown test sample, the supervised learning problem is to accurately determine the facial expression class to which the test sample belongs using the knowledge of the known class memberships of each image from a set of training images. The solution to this problem building an effective classifier to recognize the facial expression is hinged on the availability of representative training data. To date, much of the research in the area of Facial Expression Recognition (FER) is still based on posed (acted) facial expression databases, which are often exaggerated and therefore not representative of real life affective displays, as such there is a need for more publically accessible spontaneous databases that are well labelled. This thesis therefore reports on the development of the newly collected Loughborough University Spontaneous Expression Database (LUSED); designed to bolster the development of new recognition systems and to provide a benchmark for researchers to compare results with more natural expression classes than most existing databases. To collect the database, an experiment was set up where volunteers were discretely videotaped while they watched a selection of emotion inducing video clips. The utility of the new LUSED dataset is validated using both traditional and more recent pattern recognition techniques; (1) baseline results are presented using the combination of Principal Component Analysis (PCA), Fisher Linear Discriminant Analysis (FLDA) and their kernel variants Kernel Principal Component Analysis (KPCA), Kernel Fisher Discriminant Analysis (KFDA) with a Nearest Neighbour-based classifier. These results are compared to the performance of an existing natural expression database Natural Visible and Infrared Expression (NVIE) database. A scheme for the recognition of encrypted facial expression images is also presented. (2) Benchmark results are presented by combining PCA, FLDA, KPCA and KFDA with a Sparse Representation-based Classifier (SRC). A maximum accuracy of 68% was obtained recognizing five expression classes, which is comparatively better than the known maximum for a natural database; around 70% (from recognizing only three classes) obtained from NVIE.

One of the most important threats for many firms today, despite their nature of the operation, size and longevity, is insolvency. Existing empirical evidence has shown that in the past two decades, business failures have occurred at a higher rate than any time since the 1930s. Many business failure studies have been conducted over time using financial ratios as inputs and traditional statistical techniques. Some of these studies examined whether cash flow information improves the prediction of business failure. Most recently, researchers have employed discriminant analysis to perform business failure prediction. The recent changes in the world caused by unstable environments where many firms fail more than ever, there is increasing need to predict business failure. To this date, there have been limited previous studies conducted on failure prediction for UK firms. Even in other countries, there has been a small amount of research done in the field of firm failures. Therefore, this study investigates the extension of Altman’s (1968) original model in predicting the health of UK firms using discriminant analysis and performance ratios to test which ratios are statistically significant in predicting the health of the UK firms .a selected sample containing 90 failed and 1000 non failed on UK industrial firms from 2000 – 2013. The main purpose of this study is to contribute towards Altman’s (1968) original Z-score model by adding new variables (Cash flow ratio). The study found that cash flow, when combined with Altman’s original variables is highly significant in predicting the health of UK general firms. A J-UK model was developed to test the health of UK firms. When compared with the re-estimated the Altman’s original model in the UK context, the predictive power of the model was 82.9%, which is consistent with Taffler’s (1982) UK model. Furthermore, to test the predictive power of the model before, during and after the financial crisis periods; results show that J-UK model had a higher accuracy to predict the health of UK firms than the re-estimated Altman’s original model. Finally, the study proves that liquidity, profitability, leverage and capital turnover ratios are significant ratios in predicting failure. Liquidity and profitability have the highest contribution to the results of both re-estimated Altman’s original model and J-UK model. This study has implications for decision makers. Regulatory bodies and practitioners have to take into account the ratios, which contributed highest to the model in order to serve as early warning signals for corrective action.

Reliability and computational efficiency of classification error estimators are critical factors in classifier design. In a high-dimensional data setting where data is scarce, the conventional method of error estimation, cross-validation, can be very computationally expensive. In this thesis, we consider a particular discriminant analysis type classifier, the Randomly-Projected RLDA ensemble classifier, which operates under the assumption of such a ‘small sample’ regime. We conduct an asymptotic study of the generalization error of this classifier under this regime, which necessitates the use of tools from the field of random matrix theory. The main outcome of this study is a deterministic function of the true statistics of the data and the problem dimension that approximates the generalization error well for large enough dimensions. This is demonstrated by simulation on synthetic data. The main advantage of this approach is that it is computationally efficient. It also constitutes a major step towards the construction of a consistent estimator of the error that depends on the training data and not the true statistics, and so can be applied to real data. An analogous quantity for the Randomly-Projected LDA ensemble classifier, which appears in the literature and is a special case of the former, is also derived. We motivate its use for tuning the parameter of this classifier by simulation on synthetic data.

Based on the curriculum of 2011 a new grading scale ranging from A-F was introduced in the Swedish upper secondary school system. Previous research on similar data have focused on the earlier grading scale, and its crucial that the new circumstances are addressed to understand the impact on grading. Using 348 essays from the national writing test this study investigates the use of automated essay scoring as a way of grading in this new setting. Using various classication methods the models for younger students outperform the corresponding models for older students. This implies that it is harder to predict grades on essays written by older students. Based on the current data the result shows that with the new grading scale the use of automated essay scoring should be used with caution.

Este trabalho explora a contribuição da segmentação de dados, manual e estatística, combinada com análise discriminante e com redes neurais, para a tomada de decisão de concessão de crédito ao consumidor. A grande importância que a decisão de concessão de crédito tem para o mercado varejista e para a área de controladoria de uma empresa dão cenário para o aumento da relevância do gerenciamento do risco de crédito. O mercado necessita, cada vez mais, de modelos capazes de produzir boas expectativas do comportamento dos clientes, com vistas de reduzir perdas com inadimplência. Dado um banco de dados composto por 50 mil clientes de uma importante loja do setor varejista, primeiro aplica-se a análise discriminante, depois as redes neurais, para que se classifique a capacidade preditiva de cada técnica nesta etapa. Posteriormente, os dados são segmentados com base na região à qual a filial de venda pertence e, depois, por meio das análises de clusters K-Means e TwoStep Cluster. A próxima etapa compreende a aplicação da análise discriminante, depois das redes neurais, para cada um dos grupos formados, tanto pela segregação por região, quanto pela segregação por meio das técnicas de análise de clusters. A última etapa abrange a comparação da soma dos acertos dos bons e dos maus pagadores obtida tanto para análise discriminante, quanto para redes neurais, combinadas com a segmentação de dados, com os resultados obtidos na primeira etapa sem a segmentação dos dados. O modelo híbrido que combina a segmentação manual dos dados com análise discriminante e com redes neurais, formando-se 21 micro-regiões foi o que apresentou maiores porcentagens de acerto de classificação. O modelo híbrido que combina análise discriminante e redes neurais com a análise de clusters TwoStep Cluster não apresenta resultados de classificação adequados à proposta deste trabalho, devendo, portanto, ser descartado. / This paper explores the contribution of data segmentation, and statistical manual, combined with discriminant analysis and neural networks, for making the decision to grant credit to consumers. The great importance that the decision to grant credit is for the retail market and the area of controlling a business scenario to give increasing importance of managing credit risk. The market needs, increasingly, models capable of producing good expectations of customer behavior, in order to reduce losses from default. Given a database consisting of 50 000 customers of a major retail store, the first applies to discriminant analysis, then the neural networks, in order to classify the predictive ability of each technique in this step. Subsequently, the data are segmented based on the region to which the branch belongs to sell and then through the analysis of clusters K-Means and TwoStep Cluster. The next step involves the application of discriminant analysis, neural networks then, for each of the groups formed by both the segregation by region, by segregation and by the techniques of cluster analysis. The last step includes comparing the sum of the hits of the good and bad debtors obtained for both discriminant analysis and neural networks, combined with the segmentation of data, with the results obtained in the first stage - without the segmentation of the data. The hybrid model that combines the manual segmentation of the data with discriminant analysis and neural networks, forming 21 micro-regions showed the highest percentage of correct classification. The hybrid model that combines neural networks and discriminant analysis with cluster analysis results TwoStep Cluster does not have appropriate rating to the proposal of this work and should therefore be discarded.

Neste trabalho estuda-se a aplicação de métodos estatísticos multivariáveis para a classificação de gasolinas comerciais em conformidade à legislação vigente. Atualmente, a ANP baseia a classificação em limites máximos e mínimos para uma série de diferentes propriedades físico-químicas. O objetivo do trabalho é propor uma metodologia para fazer uma triagem das amostras coletadas durante o Programa de Monitoramento da Qualidade dos Combustíveis através de um método de classificação. Ela utiliza a espectroscopia NIR, que é uma técnica rápida e não destrutiva, como método analítico. Com isto será possível reduzir o número de ensaios físico-químicos que não necessariamente seriam realizados sistematicamente em todas as amostras, reduzindo-se os custos e aumentando-se a quantidade de postos monitorados. As análises NIR produzem grandes quantidades de dados, o que leva à utilização de técnicas estatísticas multivariáveis para estabelecer as metodologias de classificação. Neste trabalho utilizam-se técnicas já consagradas, como a PCA e a PLS para a compressão dos dados e a LDA e QDA para a classificação das amostras. Os dados analisados correspondem às propriedades físico-químicas e aos espectros NIR de um conjunto de 216 amostras de gasolinas comerciais, utilizado para a concepção dos modelos de classificação, e de outro de 50 amostras, utilizado para a validação dos modelos. Os modelos testados no trabalho foram as combinações da PCA-LDA, PCA-QDA, PLS-LDA, PLS-QDA, PLS (regressão) e a análise dos gráficos de scores (biplot). Os melhores desempenhos foram obtidos pelos gráficos dos scores, em seguida pela regressão PLS, PLS-QDA, PCA-QDA e PLS-QDA. Existem ainda algumas etapas a serem alcançadas para tornar prática a utilização da classificação de gasolinas comerciais através de NIR, no entanto, a contribuição deste estudo é importante pois permitiu demonstrar a sua viabilidade técnica. / In this work, the application of multivariable statistical methods for the classification of commercial gasoline in accordance to applicable laws in Brazil is studied. In the present, the ANP bases the classification of gasoline on lower and upper bounds defined for a number of physico-chemical properties. The objective of this work is to propose an alternative analysis methodology, that is adequate for making a pre-sorting of the samples collected by the Fuel Quality Monitoring Program through a classification method. This method is based on NIR spectroscopy, that is a fast and non-destructive technique, as the analytical method. In this way, it would be possible to reduce the number of physico-chemical analyses, as it would be possible not to perform them on every sample, reducing costs and increasing the quantity and frequency of gas stations that could be monitored. NIR analyses produce a great quantity of data, that makes the use of multivariable statistical techniques necessary in order to set up classification methodologies. In this work the well-known PCA and PLS techniques are used for data compression, and LDA and QDA analyses for sample classification. The data studied correspond to the physico-chemical properties and NIR spectra of a total of 216 commercial gasoline samples, used for model design, and of a 50 samples, used for validation. The classification methods that are tested are combinations of PCA-LDA, PCA-QDA, PLS-LDA, PLS-QDA, PLS (regression) and data compression scores graphical analysis (biplot). Best performance was obtained with compression scores graphical analysis, followed by PLS regression, PLS-QDA, PCA-QDA and PLS-QDA. There are still some steps to be fulfilled before the usage of commercial gasoline classification through NIR could be practical. However, this study has shown that this methodology is technically feasible.

O trabalho aqui apresentado é uma análise retrospectiva de modelos desenvolvidos, no Brasil, sobre o estudo da previsão de insolvência das empresas, objetivando-se avaliar a aplicação de métodos quantitativos para fins de análise de demonstrações contábeis. Considera-se que é relevante a avaliação da continuidade da empresa, e que, se for possível identificar fato em contrário, o uso de modelos de previsão é de importância no que tange à decisão de concessão de crédito, tanto no âmbito da intermediação financeira, realizada pelos bancos, quanto no âmbito de transações comerciais entre fornecedores e clientes. Desta última, pode-se inferir sobre a avaliação da concessão ou não da Concordata para uma empresa, servindo aqueles modelos como ferramental de análise da capacidade da empresa em cumprir o acordo da concordata, ponto esse explorado nesta pesquisa. Através da aplicação dos modelos sobre uma amostra de empresas que haviam solicitado a concordata, pôde-se avaliar se mantinham uma capacidade de discriminar as empresas que lograriam êxito na concordata. Como ferramental estatístico, é utilizada a Análise Discriminante, técnica de análise multivariada, que busca classificar os dados em dois grupos específicos. Neste trabalho, foram definidos como grupo de empresas solventes e grupo de empresas insolventes. Verificou-se que as premissas para utilização da técnica estatística de Análise Discriminante podem limitar, não invalidar, esses modelos. Há necessidade de se avaliarem os dados das amostras para se verificar se é possível ou não o uso da técnica de Análise Discriminante, além do que necessitam recorrentemente, de ser recalculados. Essa limitação reduziu-se quando se utilizaram os modelos em conjunto ou integrados, como verificado nos testes realizados. Outra técnica utilizada nesse estudo foi a de se gerar um modelo que congregue os melhores indicadores dos modelos analisados, obtendo-se um modelo de previsão, que pode ser considerado híbrido ou misto. Esse modelo foi testado quanto à sua capacidade de avaliar se as empresas concluiriam suas concordatas e, também, em sua capacidade de discriminar as empresas nos dois grupos anteriormente descritos (Solventes e Insolventes), ambos formados por empresas situadas em Belo Horizonte, Betim e Contagem. Como ressaltado, existem limitações ao uso desses modelos, que se iniciam pela própria ferramenta da Análise Discriminante. Porém, a sua utilização pode tornar mais objetiva a decisão de se conceder ou não a Concordata a uma empresa, ou, até mesmo, uma linha de crédito especial para cliente de um fornecedor ou de uma instituição bancária que se encontre nessa situação. Portanto, verificou-se ser possível, através das demonstrações contábeis das empresas objeto do estudo, a previsão da tendência de solvência ou insolvência daquelas, avaliando-se se lograriam êxito com a concordata. / This study is a retrospective analysis of models developed in Brazil with respect to the study of forecasting company insolvency, aimed at evaluating the application of quantitative methods to the financial analysis of financial statements. Evaluating the going-concern of companies is considered relevant. If facts can be identified indicating the opposite, the use of forecasting models is important what the decision on the extension of credit is concerned, not only in the field of financial intermediation, realized by banks, but also in the field of commercial transactions between suppliers and clients. From this decision, inferences can be made about the evaluation of whether a composition of debt will be conceded to a company, in which the models mentioned above will serve as tools for analyzing the company’s capacity to fulfill the composition agreement, an issue that is dealt with in this research. By means of the application of those models to a sample of companies that had applied for composition of debt, it could be evaluated whether the models maintained their capacity to distinguish the companies that were successful in the composition of debt. As a statistical tool, the Discriminant Analysis is used. This is a multivariate analysis technique that seeks to classify the data in two specific groups. In this study, they were defined as solvent companies group and insolvent companies group. It was verified that the premises for using the statistical technique of Discriminant Analysis can limit, but not invalidate these models. The data of the samples need to be assessed in order to verify whether it is possible or not to use the Discriminant Analysis technique. In addition, they recurrently need to be recalculated. This limitation was reduced when the models were used together or in an integrated way, as verified in the accomplished tests. Another technique used in this study was the creation of a model that unites the best indicators of the models that were analyzed, obtaining a forecasting model, which can be considered a hybrid or mixed. This model was tested for its capacity to evaluate whether the companies would conclude the composition of debt as well as its capacity to discriminate the companies in the two groups previously described (Solvent and Insolvent), both of which consist of companies located in Belo Horizonte, Betim and Contagem. As highlighted, the use of these models is limited, starting with the Discriminant Analysis tool itself. Nevertheless, their utilization can make the decision on the concession of debt composition to a company more objective, or even the decision on extending a special credit line to the customer of a supplier or to the client of a bank who finds himself in this situation. Therefore, it was confirmed that the analysis of the financial statements of the firms included in this study permits to forecast the possibility to determine the solvency or insolvency trend of the firms, as well as to assess their eventual success with the concordat.

Novas regiões vitícolas estão se estabelecendo em muitas regiões brasileiras. Em especial, no estado do Rio Grande do Sul diversos empreendimentos estão instalados ou em fase de projeto de instalação em várias regiões. No contexto da moderna vitivinicultura nacional, um fator importante de agregação de valor aos produtos resultantes da atividade é a sua associação a uma identidade regional, tendência próxima ao conceito de terroir. Tratando-se de regiões produtoras novas, faz-se necessário um complexo levantamento de características de cada região produtora, de modo que sejam definidos parâmetros de diferenciação que confiram, a cada região, a necessária tipicidade tão próxima à ideia de terroir. Este trabalho propõe-se a trazer uma contribuição à caracterização de algumas novas regiões vitícolas do Rio Grande do Sul, tentando mostrar que é possível evidenciar fatores físicos que diferenciam cada região. Foram escolhidas três áreas na metade sul do estado, sendo duas na Campanha Gaúcha (Almadén em Santana do Livramento e Seival em Candiota), e uma na Serra do Sudeste (Chandon em Encluzilhada do Sul); também foi estudada uma área na Serra Gaúcha (Boscato em Nova Pádua). Como ferramentas de estudo, foram utilizadas técnicas de espectrorradiometria para levantamento de dados de campo nas quatro regiões, onde foram tomados espectros de reflectância foliar no visível e no infravermelho próximo e médio. Foram selecionadas parcelas de parreirais das variedades Cabernet Sauvignon, Merlot, Pinot Noir, Chardonnay e Riesling Itálico. Também foram utilizadas imagens de satélite (ASTER) para estudar as áreas da Almadén e Seival, com dados de reflectância no visível e no infravermelho de diversos cultivares de Vitis vinifera. Os dados foram analisados usando-se diversas técnicas de separação, como algoritmos de classificação supervisionada e Análise Discriminante. Os resultados, tanto para os dados de radiometria de campo quanto para os dados orbitais, mostraram que a partir da reflectância de folhas e de dossel é possível separar cada uma das regiões, com acurácias da ordem de 80% ou mais, sendo esta separabilidade atribuída à influência do meio físico sobre as plantas. Conclui-se que a utilização de dados e técnicas de Sensoriamento Remoto, com o apoio de técnicas de análise estatística, constitui relevante ferramenta de apoio à caracterização de regiões vitícolas no Rio Grande do Sul, e provavelmente para qualquer região produtora. / New viticultural regions are being created in several regions across Brazil. For instance, in Rio Grande do Sul State many wineries are already established, are being installed or are in project phase. In the context of modern Brazilian viticulture, an important factor for added value to products from the activity is its association to a regional identity, a tendency which is akin to the terroir concept. Being new regions, a comprehensive survey of the characteristics of each producing area is necessary, a step leading to the definition of parameters of differentiation, which will give to each region the required typicity, crucial to the terroir idea. Presently, we carry out a contribution to the characterization of some new viticultural regions of Rio Grande do Sul, trying to show that it is possible to bring to light physical factors which will differentiate each region. Three areas in the State’s Metade Sul (southern half) were selected, being two in the Campanha Gaúcha region (Almadén in Santana do Livramento and Seival in Candiota), and another one at the Serra do Sudeste (Chandon in Encruzilhada do Sul); we also selected a winery at the Serra Gaúcha (Boscato in Nova Pádua). As tools for this study, we used techniques of spectroradiometry to collect field data in all four regions, acquiring spectra of leaf reflectance in visible, wavelengths, and at near and mean infrared as well. We selected vineyards of the grape varieties Cabernet Sauvignon, Merlot, Pinot Noir, Chardonnay and Riesling Itálico. Satellite images (ASTER product) were also taken to study the Almadén and Seival wineries, using reflectance data in visible and infrared for some varieties of Vitis vinifera. All data were analyzed through several techniques intended for differentiation, as algorithms for supervised classification and, in Statistics, Discriminant Analysis. The results, from radiometry field data and from satellite data as well, showed that from the reflectance of leaves and canopy it is possible to separate each region, with accuracies as high as 80% and even more. This separability is believed to be due to the influence of the physical environment on plants. It is concluded that the use of data and techniques from Remote Sensing, associated with techniques of statistical analysis, are relevant tools to support the characterization of viticultural regions in Rio Grande do Sul and probably in any producing region.

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

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.