O objetivo geral desta pesquisa foi investigar as contribuições das ouvidorias para as atividades da gestão e dos órgãos do sistema de governança corporativa das instituições bancárias, baseado na regulamentação brasileira. A fundamentação teórica contemplou as ouvidorias, as temáticas da governança corporativa e da responsabilidade social corporativa. Realizou-se um estudo descritivo e exploratório, com abordagem qualitativa e quantitativa, segmentado em três etapas. Na primeira etapa, efetuou-se uma pesquisa de campo com coleta por questionário da opinião de 24 especialistas, possibilitando a proposição de um quadro analítico com 2 itens e 19 subitens emergentes da fundamentação teórica para caracterizar aspectos da gestão e da governança corporativa pertinentes às ouvidorias. Na segunda etapa, empregou-se o quadro proposto para analisar o conteúdo da documentação pertinente a 26 instituições bancárias da amostra intencional, selecionada por acessibilidade aos dados do 1º semestre de 2008 ao 1º semestre de 2011. Na terceira etapa, demonstrou-se a evolução dos assuntos de 1.377 reclamações pertinentes às ouvidorias das instituições da amostra, divulgadas pelo Banco Central do Brasil. Para o tratamento dos dados, foram empregadas a estatística descritiva, a estatística inferencial, com abordagem não paramétrica predominante, e a Análise de Correspondência Múltipla. Os principais resultados mostraram que o percentual de subitens presentes para as características de gestão foi 67,09% e para as características de governança corporativa foi 70,77%. Não houve diferenças significativas entre os referidos percentuais, mas foram constatadas diferenças ao segmentar tais instituições pelo porte, pelo tipo de controle e pelo tipo de listagem. Quanto às contribuições das ouvidorias para as atividades da gestão, foi divulgado que a ouvidoria implementa o atendimento em última instância e o tratamento de reclamações em 100% das instituições. Todavia, apurou-se, no período estudado, que instituições de porte muito grande em clientela e com controle público poderiam ter aprimorado o atendimento da ouvidoria, e instituições de porte médio e privadas nacionais poderiam ter aprimorado, entre outros assuntos, o prazo de resposta da ouvidoria. Foi divulgado que a ouvidoria pode propor melhorias de procedimentos e rotinas em todos os casos examinados, mas as proposições de melhoria não são respaldadas, apenas, em análises das reclamações recebidas de clientes e usuários, e podem ocorrer antes da oferta do produto. Quanto às contribuições das ouvidorias para as atividades dos órgãos do sistema de governança corporativa, houve evidências do reporte a respeito das ouvidorias para o conselho de administração, para auditoria interna, para comitê de auditoria e para auditoria independente, mas constatou-se, em geral, que os percentuais de divulgação das responsabilidades desses órgãos a respeito das informações reportadas pode ser aperfeiçoado. Este estudo apresenta uma aplicação original de técnica multivariada para auxiliar a interpretação de dados qualitativos a respeito das ocorrências de reclamações pertinentes às ouvidorias no setor bancário brasileiro, pode contribuir para a avaliação de políticas pertinentes à implementação das atribuições dessas ouvidorias que levem em conta variáveis de governança corporativa baseadas em práticas recomendadas pelo IBGC, pode colaborar para o entendimento da contribuição das ouvidorias para prevenção de riscos e pode subsidiar reflexões de clientes, usuários e demais partes interessadas na participação das ouvidorias, e dos órgãos do sistema de governança corporativa, em processos decisórios das instituições bancárias no Brasil. / The general aim of this research was to investigate the contributions of the ombudsman\'s offices for the activities of management and of the organs of corporate governance of banks, based on Brazilian regulation. The theoretical framework covered the ombudsman\'s offices, issues of corporate governance and corporate social responsibility. A descriptive and exploratory study was carried out with a qualitative and quantitative approach, divided into three stages. The first step was field survey with data collected by questionnaire on the opinion of 24 experts, which enabled an analytical framework to be constructed with 2 items and 19 sub-items emerging from the theoretical framework to characterize aspects of management and corporate governance relevant to the ombudsman\'s offices. In Step 2 we used the proposed framework to analyze the content of the documents relevant to 26 banks in the intentional sample, selected by accessibility to data from the 1st semester of 2008 to the 1st semester of 2011. Step 3 showed the evolution of issues of 1.377 complaints relevant to the ombudsman\'s offices for the institutions in the sample, provided by the Central Bank of Brazil. The data were processed using descriptive statistics and inferential statistics, with a predominantly non-parametric approach, and Multiple Correspondence Analysis. The main results showed that the percentage of sub-items present for the management features was 67.09% and for corporate governance features was 70.77%. There were no significant differences between these percentages, but differences were found when the institutions were divided by size, by type of control and type of listing. Regarding the contributions of the ombudsman\'s offices for management activities, it was reported that the ombudsman\'s office implements the last resort assistance and the complaint handling in 100% of the institutions. However, it was found in the period studied that institutions with a very large clientele and public control could have improved the ombudsman\'s office assistance, and mid-sized and private national institutions could improve, among other things, the response time of the ombudsman\'s office. It was reported that the ombudsman\'s office may propose improvements of procedures and routines in all cases examined, but the improvement proposals are not supported only in analyses of complaints received from customers and users, and they may occur before the product offering. With regard to the contributions of the ombudsman\'s offices to the activities of the bodies of the corporate governance system, there was evidence of reporting in relation to the ombudsman\'s offices to the board of directors, internal audit, audit committee and to the independent auditor, but it was found, in general, that the percentage of disclosure of the responsibilities of these bodies regarding the information reported could be improved. This study presents an original application of multivariate techniques to assist the interpretation of qualitative data about the occurrence of complaints relevant to the ombudsman\'s offices in the Brazilian banking sector, and can contribute to the evaluation of policies relevant to the implementation of the duties of these ombudsman\'s offices taking into account corporate governance variables based on best practices recommended by the IBGC. It may also contribute to understanding the contribution of the ombudsman\'s offices to the prevention of risk and can support reflections of customers, users and other stakeholders in the participation of ombudsman\'s offices and the bodies of the corporate governance system in the decision-making processes of banks in Brazil.

This paper investigates variation of lexical and analytic causatives in 15 European languages from the Germanic, Romance, and Slavic genera based on a multilingual parallel corpus of film subtitles. Using typological parameters of variation of causatives from the literature, this study tests which parameters are relevant for the choice between analytic and lexical causatives in the sample of languages. The main research question is whether the variation is constrained by one semantic dimension, namely, the conceptual integration of the causing and caused events, as suggested by previous research on iconicity in language, or whether several different semantic and syntactic factors are at play. To answer this question, I use an exploratory multivariate technique for categorical data (Multiple Correspondence Analysis with supplementary points) and conditional random forests, a nonparametric regression and classification method. The study demonstrates the importance of corpus data in testing typological hypotheses.

Η παρούσα διπλωματική εργασία ασχολείται με πολυδιάστατα κατηγορικά δεδομένα όπως αυτά προκύπτουν από συλλογή μέσω ερωτηματολογίων. Για να αναλυθεί όμως ένα ερωτηματολόγιο το οποίο περιλαμβάνει πλήθος ερωτήσεων-μεταβλητών και να εξάγουμε ορισμένα χρήσιμα συμπεράσματα θα πρέπει, πρώτα από όλα να γίνει η κατάλληλη κωδικοποίηση των δεδομένων. Χρησιμοποιώντας στατιστικές τεχνικές και μεθόδους κατάλληλες για κατηγορικά δεδομένα μπορούμε πιο εύκολα να μελετήσουμε τις σχέσεις μεταξύ των μεταβλητών. Για τον σκοπό αυτό, παρουσιάζουμε και αναλύουμε τη θεωρία της Παραγοντικής Ανάλυσης των Αντιστοιχιών και της Πολλαπλής Παραγοντικής Ανάλυσης των Αντιστοιχιών. Ύστερα, αναλύουμε τα αποτελέσματα που δίνουν οι μέθοδοι όταν τις χρησιμοποιήσουμε για την ανάλυση του ερωτηματολογίου. Τα συμπεράσματα που προκύπτουν έχουν ιδιαίτερο ενδιαφέρον. Η δυναμικότητα των μοντέλων αυτών παρουσιάζεται μέσα από μία εφαρμογή από τον χώρο των κοινωνικών επιστημών σε θέματα που αφορούν τα σχολεία και τους μαθητές της Πρωτοβάθμιας εκπαίδευσης. Στις μεθόδους αυτές δίνεται έμφαση κυρίως στα γραφικά αποτελέσματα αλλά και στις εκτιμήσεις των σκορ των κατηγοριών των μεταβλητών. Όλα τα παραπάνω τα συγκρίνουμε κριτικά μεταξύ τους στη θεωρία και στη πράξη έτσι ώστε ο ενδιαφερόμενος αναγνώστης να κατανοήσει περισσότερο τις μεθόδους αυτές και να αποκομίσει όσο το δυνατόν περισσότερες πληροφορίες που θα τον βοηθήσουν για την εφαρμογή τους. / This dissertation deals with multivariate categorical data of a raw data set produced by a questionnaire designed for a research purpose. However, in order to analyze a questionnaire and extract some fruitful results, that includes a great number of questions-variables, we must first impose a structure on it especially on situations this specific structure is missing. Whenever the structure is imposed, by using statistical techniques and methods designed for categorical data, we can then study more efficiently the relations among the variables in concern for further analyses. The capacity of these models is presented through an application from the social sciences on issues concerning schools and pupils in primary education. The analysis on a smaller subset is further explored by describing the issues of Correspondence Analysis, and Multiple Correspondence Analysis. With these methods, we focus on the interpretation of the results on the graphical displays of the data but also on the estimated category scores of the variables. The above methods described in this dissertation and the results after implementing them are all critically compared with each other at each chapter. This gives to the interesting reader the possibility to fully understand them and to obtain additional information on their implementation.

Submitted by Franciele Moreira (francielemoreyra@gmail.com) on 2018-01-31T14:05:36Z No. of bitstreams: 2 Tese - Elis Regina de Olivieira - 2017.pdf: 5392456 bytes, checksum: 68610e861c2d0e531893fa02a36401f5 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Approved for entry into archive by Luciana Ferreira (lucgeral@gmail.com) on 2018-01-31T14:28:36Z (GMT) No. of bitstreams: 2 Tese - Elis Regina de Olivieira - 2017.pdf: 5392456 bytes, checksum: 68610e861c2d0e531893fa02a36401f5 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Made available in DSpace on 2018-01-31T14:28:36Z (GMT). No. of bitstreams: 2 Tese - Elis Regina de Olivieira - 2017.pdf: 5392456 bytes, checksum: 68610e861c2d0e531893fa02a36401f5 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Previous issue date: 2017-12-15 / Fundação de Amparo à Pesquisa do Estado de Goiás - FAPEG / This thesis identified and analyzed the factors that determine the decision- making process of beef producers in relation to their investments in technology and the environmental consequences of these decisions, focusing on the Vermelho River basin as the geographical unit of analysis. The study analyzed the productive and economic viability of four models of beef production based on grazing systems, considering distinct levels of technology. This study defined three analytical dimensions: the socio- economic profile of the producer, the configuration of the production system, and management practices. The research used a stratified sampling design, with 60 randomly-selected cattle ranching establishments distributed in the strata according to their relief and the variability of their soils. The responses to the standard questionnaire provided primarily qualitative data, with the variables being measured on an ordinal or nominal scale. Multiple correspondence and cluster analyses were used to identify, respectively, the components that contributed most to the variance observed in the analyses, and the groups of variables and sampling units which best represented the indices of technology and pasture degradation, with a beta regression being applied to determine a mathematical function that could be applied to infer the classification of other sampling units in the study basin. The technological and pasture degradation indices did not present any spatial autocorrelation, based on Moran’s global or local indices. Nonparametric statistics (Fisher’s exact test, Spearman and Kendall) were used to verify the significance of the associations among the variables tested during the study. Principal results: the system with the lowest input of supplements was the most profitable; no significant relationship was found between the technological index and any of the variables: education, farm credit, cattle density, pasture degradation index, or the interaction of this index with the size of the property. Significant associations were identified between the technological index and the income, relief, and size of the property. The positive association between income and technology indicates that more technological systems are more profitable. The inverse relationship between relief and technology reflects the fact that the more technological systems tend to be installed on flatter or gently sloping terrain. The direct association between the size of the property and the use of technology reflects the importance of obtaining economies of scale to guarantee the profitability of the investment. The lack of any systematic relationship between the technological index and cattle density, as confirmed by the cluster analysis, indicates that high densities (above 2 animals per hectare) may be found in low technology systems, which impacts productivity and profitability, as well as long-term environmental sustainability. / Esta tese tem por objetivo identificar e analisar os fatores que condicionam a tomada de decisões dos produtores de bovinos por inversão de capital em níveis de tecnificação e suas consequências ambientais, utilizando a Bacia Hidrográfica do Rio Vermelho como unidade geográfica de estudo. Analisou-se a viabilidade produtiva e econômica de quatro modelos de produção de bovinos de corte em sistema de pastejo, considerando opções tecnológicas distintas. A partir desse estudo o instrumento de coleta de dados foi estruturado em três dimensões: perfil socioeconômico do gestor; composição do sistema de produção; e manejos. A pesquisa utilizou amostra estratificada por compartimentos com 60 estabelecimentos agropecuários, distribuídos entre os estratos, conforme a declividade e variabilidade de solos, selecionadas aleatoriamente. As perguntas e respostas fechadas constantes do formulário resultaram principalmente em variáveis de natureza qualitativa, com respectivas categorias medidas principalmente em escala nominal ou ordinal. Utilizou-se análise de correspondência múltipla e análise de cluster para identificar as componentes com maior proporção de variância explicada, e os respectivos agrupamentos de variáveis e unidades amostrais, que representassem melhor os indicadores de nível tecnológico e de degradação de pastagens, utilizando regressão beta para determinar equação matemática que possibilitasse inferir essa classificação para outras unidades amostrais da bacia. Indicador tecnológico e de degradação de pastagens não apresentaram autocorrelação espacial, considerando Índice Moran Global e Local. Utilizou-se de testes não paramétricos (teste exato de Fisher, Spearman e Kendall) para verificar a existência de associação entre as variáveis que compuseram as hipóteses científicas. Principais resultados: sistema de baixo consumo de suplemento apresentou maior índice de rentabilidade; não foi observada evidência estatística entre indicador tecnológico com cada uma das variáveis: educação, crédito rural, taxa de lotação, indicador de degradação de pastagens; e entre degradação de pastagens e tamanho da propriedade. As associações verificadas foram entre indicador tecnológico com cada uma das variáveis: renda, relevo e tamanho da propriedade. Renda associada positivamente com tecnologia sugere que, sistemas mais tecnificados remuneram melhor o gestor. Relevo, associado inversamente com tecnologia, revela que sistemas mais tecnificados estão instalados em relevos planos e em suave ondulados. Associação direta entre tamanho da propriedade e tecnologia evidencia que sistemas tecnificados de produção de bovinos em pastejo, demandam maior área para obter ganho de escala. A não associação entre indicador tecnológico e taxa de lotação, conforme observado também pela análise de cluster, sugere que taxas de lotação mais elevadas (acima de 2 UA/ha) podem ocorrer em sistemas com baixa tecnificação, comprometendo a sustentabilidade produtiva, econômica e ambiental.

Ensemble models have proved effective in a variety of classification tasks. These models combine the predictions of several base models to achieve higher out-of-sample classification accuracy than the base models. Base models are typically trained using different subsets of training examples and input features. Ensemble classifiers are particularly effective when their constituent base models are diverse in terms of their prediction accuracy in different regions of the feature space. This dissertation investigated methods for combining ensemble models, treating them as base models. The goal is to develop a strategy for combining ensemble classifiers that results in higher classification accuracy than the constituent ensemble models. Three of the best performing tree-based ensemble methods – random forest, extremely randomized tree, and eXtreme gradient boosting model – were used to generate a set of base models. Outputs from classifiers generated by these methods were then combined to create an ensemble classifier. This dissertation systematically investigated methods for (1) selecting a set of diverse base models, and (2) combining the selected base models. The methods were evaluated using public domain data sets which have been extensively used for benchmarking classification models. The research established that applying random forest as the final ensemble method to integrate selected base models and factor scores of multiple correspondence analysis turned out to be the best ensemble approach.

Diploma thesis is focused on social stratification aspects of internet use, as a part of leisure time activities and consequently as a potential part of cultural capital. The main objective of this paper is to make a link between Bourdieusian tradition of research of cultural capital and research tradition of stratification aspects of internet use (especially digital divide). The author is trying to prove if the type of internet usage could be considered as a part of cultural capital. In the first, theoretical part of this text, the author is presenting both research traditions concerned, trying to synthetize them and concluding that internet usage can be considered as a capital. In the second, analytical part of this paper, the author is using data from World Internet Project to provide, that the theoretical synthesis have some empirical support. The analysis proves that the structure of Czech internet users' activities are differentiated by their cultural capital level. On the basis of this analysis, we can conclude that more powerful differentiation between types of internet usages is provided by traditional form of cultural capital, rather than by the emerging one. Powered by TCPDF (www.tcpdf.org)

Thanks to the advanced technologies and social networks that allow the data to be widely shared among the Internet, there is an explosion of pervasive multimedia data, generating high demands of multimedia services and applications in various areas for people to easily access and manage multimedia data. Towards such demands, multimedia big data analysis has become an emerging hot topic in both industry and academia, which ranges from basic infrastructure, management, search, and mining to security, privacy, and applications. Within the scope of this dissertation, a multimedia big data analysis framework is proposed for semantic information management and retrieval with a focus on rare event detection in videos. The proposed framework is able to explore hidden semantic feature groups in multimedia data and incorporate temporal semantics, especially for video event detection. First, a hierarchical semantic data representation is presented to alleviate the semantic gap issue, and the Hidden Coherent Feature Group (HCFG) analysis method is proposed to capture the correlation between features and separate the original feature set into semantic groups, seamlessly integrating multimedia data in multiple modalities. Next, an Importance Factor based Temporal Multiple Correspondence Analysis (i.e., IF-TMCA) approach is presented for effective event detection. Specifically, the HCFG algorithm is integrated with the Hierarchical Information Gain Analysis (HIGA) method to generate the Importance Factor (IF) for producing the initial detection results. Then, the TMCA algorithm is proposed to efficiently incorporate temporal semantics for re-ranking and improving the final performance. At last, a sampling-based ensemble learning mechanism is applied to further accommodate the imbalanced datasets. In addition to the multimedia semantic representation and class imbalance problems, lack of organization is another critical issue for multimedia big data analysis. In this framework, an affinity propagation-based summarization method is also proposed to transform the unorganized data into a better structure with clean and well-organized information. The whole framework has been thoroughly evaluated across multiple domains, such as soccer goal event detection and disaster information management.

The aim of this thesis is to introduce statistical method called Correspondence analysis as a strong instrument for exploratory data analysis. The main purpose is to understand how to interpret the correspondence map, the graphical output of this method, correctly. The method is presented both in its simple version, and its extension to multivariate data. Usage of method is demonstrated on data of non-financial subjects of Czech republic, which are entered in the register of economic subjects.

No description available.

This study analyzes, the structure of the Erasmus Mundus scholarship program by utilizing mixed methods, and is being analyzed within the contextualized framework of the Cambodia’s Higher Education system as well as its relationship with the Erasmus Mundus scholarship program. Mixed approach has been employed for this study, combining both quantitative data to construct the social space, which acts as a backbone for interpretation, and qualitative data from interviewing the scholarship program coordinator and scholarship holders in order to investigate the recruitment process and the students’ perception respectively, presenting the macro-micro relationship that makes up the entirety of the scholarship program. Collectively inspected, the findings reveal a new contextualized result that contributes not only to the development of the Cambodia’s Higher Education, but also to its position and its students within the space of the scholarship program of Erasmus Mundus. Through the close inspection using interview method, the underlying complexity of the recruitment process of the scholarship program is illustrated, indicating a multi-layer hierarchy and multiple decision-making processes. There are also implications of oppositions between the properties of the students, signaling the diversified student body in the scholarship program. Prominent theme of the opposition in the space appears to be related to the colonial past of the Cambodia’s system as well as the differences between local and international experiences that the students possess. Students interviewed in the study, furthermore, reveal how they prepared themselves to apply for the scholarship program and to be successful candidate, which depicts their assets and strategies that can be derived from the constructed space. Unanimously the students show strong confidence and great insight of information which can be understood as the compatibilities between the recruiter of the scholarship program and the students’ assets. Finally, overall findings give a new perspective of the profile of the scholarship program within the Erasmus Mundus mobility scheme. It shows that while students are required to possess certain objectively set requirements, there are also exception cases that students who possessed lower amount of capitals can also be successful, provided that the conditions like the credentials of their degree from Cambodia, as well as minimum requirements to pass the university admission have been met, and their “motivation letter” is exceptionally well-written which shows the subjective side of the recruitment process.