In dieser Arbeit werden quantenkritische Phänomene der ferromagnetisch stark korrelierten Schwere-Fermionen-Systeme YbRh2Si2, YbNi4P2 und CeFePO untersucht. Hierzu sind Messungen des elektrischen Widerstands und der AC-Suszeptibilität durchgeführt worden. Das System YbRh2Si2 besitzt einen antiferromagnetischen Phasenübergang bei TN = 0.07 K. Die Verletzung des Wiedemann-Franz-Gesetzes an seinem Magnetfeld-induzierten quantenkritischen Punkt kann indirekt durch neue Widerstandsmessungen bestätigt werden. Mit den Substitutionen Yb(Rh1-xCox)2Si2, Yb(Rh1-yIry)2Si2 und Yb1-zLazRh2Si2 kann die Übergangstemperatur erhöht oder erniedrigt werden. Dadurch lässt sich sowohl ein weiterer quantenkritischer Punkt erreichen als auch das Verhältnis zwischen ferromagnetischen und antiferromagnetischen Korrelationen einstellen. Das durch chemische Substitution erzeugte magnetische Phasendiagramm wird detailliert untersucht. Es wird gezeigt, dass die zunehmenden ferromagnetischen Fluktuationen mit steigender Cobalt-Konzentration zu einem ferromagnetischen Phasenübergang bei x = 0.27 führen. Die magnetischen Momente ordnen entlang der magnetisch harten c-Richtung. Das neue Schwere-Fermionen-System YbNi4P2 besitzt einen, im Rahmen dieser Arbeit entdeckten, ferromagnetischen Phasenübergang bei der erstaunlich niedrigen Curie-Temperatur TC = 0.17 K. Es werden weiterführende Messungen an Einkristallen durchgeführt, die zeigen, dass die Momente senkrecht zur magnetisch weichen c-Richtung ordnen. Erste Hinweise auf einen Magnetfeld-induzierten ferromagnetischen quantenkritischen Punkt werden gefunden. Das Schwere-Fermionen-System CeFePO befindet sich in der Nähe einer ferromagnetischen Instabilität, die durch Arsen-Substitution auf dem Phosphor-Platz erreicht werden kann. Bisher ging man davon aus, dass CeFePO selbst paramagnetisch ist. In dieser Arbeit wird eine Anomalie bei T ~ 0.7 K in kürzlich hergestellten Proben als kurzreichweitige Ordnung identifiziert. Es kann gezeigt werden, dass es sich damit um eine neuartige Art und Weise handelt, wie ein ferromagnetischer quantenkritischer Punkt umgangen wird.

<p>HIV and AIDS are disease conditions that have led to high mortality rates in Southern Africa since the late 1970s. The socio-economic system has led to the unequal spread of resources&rsquo / and vulnerability and exposure to HIV is more prevalent in poorer communities. The added burden of life stresses cause for many to be isolated and stigmatised and are often not equipped with the necessary support and coping skills to deal with the magnitude of these circumstances. There is a high prevalence of mental disorders and especially depression amongst individuals infected with either HIV or AIDS. Research shows that stressful life events can impact HIV course progression and impacts the QoL of those infected with HIV or AIDS. Given the social and psychological context of HIV and AIDS, the aim of the study was to examine the relationship between depression and QoL in a sample of adults diagnosed with HIV or AIDS. This quantitative, cross-sectional study used the Becks Depression Inventory II (BDI II) and the Quality of Life Enjoyment and Satisfaction Questionnaire (Q-LES-Q), to measure the variables concerned. This battery of&nbsp / questionnaires was administered to a purposive sample of adult individuals diagnosed with HIV or AIDS residing in a previously disadvantaged area in the Cape Metropole region. Using SPSS,&nbsp / data was analysed and descriptive and inferential statistics were conducted. The study found that there were more women than men with HIV or AIDS that were found to be depressed (mild, moderate and severe depression). Furthermore, the depressive state increased when the progression of the disease increased. There were generally no significant differences in the QoL&nbsp / experienced within various areas of life and overall life satisfaction experienced. However, the QoL experienced in work was lower. There was a significant relationship between the depressed state and QoL and life satisfaction experienced in household duties and tasks. The contribution of this study includes informing the larger research project, with regards to future treatment&nbsp / regimes. It will update statistics on the prevalence of depression and QoL of adults diagnosed with HIV or AIDS in the area. This study is framed within a biopsychosocial model and is&nbsp / theoretically underpinned by Beck&rsquo / s theory of depression. Key words: HIV, AIDS, adults, depression, quality of life (QoL), Beck Depression Inventory II (BDI II), Quality of Life Enjoyment and Satisfaction Questionnaire (Q-LES-Q), Antiretrovirals (ARV&rsquo / s), prevalence data, correlations.</p>

This thesis consists of four research papers. The main focus is on building the multivariate Conditional Correlation (CC-) GARCH models. In particular, emphasis lies on considering an extension of CC-GARCH models that allow for interactions or causality in conditional variances. In the first three chapters, misspecification testing and parameter restrictions in these models are discussed. In the final chapter, a computer package for building major variants of the CC-GARCH models is presented. The first chapter contains a brief introduction to the CC-GARCH models as well as a summary of each research paper. The second chapter proposes a misspecification test for modelling of the conditional variance part of the Extended Constant CC-GARCH model. The test is designed for testing the hypothesis of no interactions in the conditional variances. If the null hypothesis is true, then the conditional variances may be described by the standard CCC-GARCH model. The test is constructed on the Lagrange Multiplier (LM) principle that only requires the estimation of the null model. Although the test is derived under the assumption of the constant conditional correlation, the simulation experiments suggest that the test is also applicable to building CC-GARCH models with changing conditional correlations. There is no asymptotic theory available for these models, which is why simulation of the test statistic in this situation has been necessary. The third chapter provides yet another misspecification test for modelling of the conditional variance component of the CC-GARCH models, whose parameters are often estimated in two steps. The estimator obtained through these two steps is a two-stage quasi-maximum likelihood estimator (2SQMLE). Taking advantage of the asymptotic results for 2SQMLE, the test considered in this chapter is formulated using the LM principle, which requires only the estimation of univariate GARCH models. It is also shown that the test statistic may be computed by using an auxiliary regression. A robust version of the new test is available through another auxiliary regression. All of this amounts to a substantial simplification in computations compared with the test proposed in the second chapter. The simulation experiments show that, under both under both Gaussian and leptokurtic innovations, as well as under changing conditional correlations, the new test has reasonable size and power properties. When modelling the conditional variance, it is necessary to keep the sequence of conditional covariance matrices positive definite almost surely for any time horizon. In the fourth chapter it is demonstrated that under certain conditions some of the parameters of the model can take negative values while the conditional covariance matrix remains positive definite almost surely. It is also shown that even in the simplest first-order vector GARCH representation, the relevant parameter space can contain negative values for some parameters, which is not possible in the univariate model. This finding makes it possible to incorporate negative volatility spillovers into the CC-GARCH framework. Many new GARCH models and misspecification testing procedures have been recently proposed in the literature. When it comes to applying these models or tests, however, there do not seem to exist many options for the users to choose from other than creating their own computer programmes. This is especially the case when one wants to apply a multivariate GARCH model. The last chapter of the thesis offers a remedy to this situation by providing a workable environment for building CC-GARCH models. The package is open source, freely available on the Internet, and designed for use in the open source statistical environment R. With this package can estimate major variants of CC-GARCH models as well as simulate data from the CC-GARCH data generating processes with multivariate normal or Student's t innovations. In addition, the package is equipped with the necessary functions for conducting diagnostic tests such as those discussed in the third chapter of this thesis. / <p>Diss. Stockholm : Handelshögskolan, 2010. Sammanfattning jämte 4 uppsatser.</p>

Business process analysis and integration are key endeavours for today's enterprises. Recently, Web services have been widely adopted for the implementation and integration of business processes within and across enterprises. In this dissertation, we investigate the problem of enabling the analysis of service interactions, in today's enterprises, in the context of business process executions, and that of service integration. Our study shows that only fraction of interactions in the enterprise are supported by process-aware systems. However, enabling above-mentioned analyses requires: (i) a model of the underlying business process to be used as a reference for the analysis, and (ii) the ability to correlate events generated during service interactions into process instances. We refer to a process model and the corresponding process instances as a "process view". We propose the concept of process space to refer to all process related information sources in the enterprise, over which various process views are defined. We propose the design and development of a system called "process space discovery system" (PSDS) for discovering process views in a process space. We introduce novel approaches for the correlation of events into process instances, focusing on the public processes of Web services (business protocols), and also for the discovery of the business protocol models from the process instances of a process view. Analysis of service integration approaches shows that while standardisation in Web services simplifies the integration in the communication level, at the higher levels of abstractions (e.g., services interfaces and protocol models) services are still open to heterogeneities. We characterise the mismatches between service interfaces and protocol specifications and introduce "mismatch patterns" to represent them. A mismatch pattern also includes an adapter template that aims at the resolution of the captured mismatch. We also propose semi-automated approaches for identifying the mismatches between interface and protocol specifications of two services. The proposed approaches have been implemented in prototype tools, and experimentally validated on synthetic and real-world datasets. The discovered process views, using PSDS, can be used to perform various analyses in an enterprise, and the proposed adaptation approach facilitates the adoption of Web services in business process integration.

Η διαδικασία της έρευνας ξεκινά με την αναγνώριση ενός προβλήματος ή ευκαιρίας της αγοράς. Υπάρχουν 3 ερωτήσεις – κλειδιά που πρέπει να απαντήσει κανείς για να βρει τη λύση στον ορισμό του παραπάνω προβλήματος: 1. Για ποιο λόγο αναζητούμε την πληροφορία; 2. Μήπως η πληροφορία υπάρχει ήδη; 3. Μπορεί πραγματικά να απαντηθεί η ερώτηση; Η δημιουργία του σχεδίου της έρευνας είναι το επόμενο στάδιο που ακολουθείται για να απαντηθούν οι αντικειμενικοί ή υποθετικοί σκοποί της έρευνας. Είναι η δόμηση ή το πλαίσιο για να λύσουμε ένα συγκεκριμένο πρόβλημα. Το εργαλείο για την επίτευξη του σχεδίου της έρευνας είναι οι Περιγραφικές Μελέτες και οι Αιτιολογικές Μελέτες. Οι πρώτες απαντούν στις ερωτήσεις ποιος, τι, που, πότε και πως; Στις αιτιολογικές μελέτες ο ερευνητής εξετάζει αν μια μεταβλητή αιτιολογεί ή προσδιορίζει την αξία μιας άλλης μεταβλητής. Μια εξαρτημένη μεταβλητή είναι μια μεταβλητή που αναμένεται να προβλεφθεί ή να εξηγηθεί. Μια ανεξάρτητη μεταβλητή είναι μια μεταβλητή σε ένα πείραμα όπου ο ερευνητής της αγοράς μπορεί, σε ένα βαθμό, να χειριστεί, να αλλάξει ή να μεταβάλλει. Υπάρχουν 3 βασικές μέθοδοι έρευνας και η επιλογή γίνεται ανάλογα με τους αντικειμενικούς στόχους του σχεδίου. Οι μέθοδοι αυτές είναι η Αξιολόγηση, η Παρατήρηση και το Πείραμα. Στη συνέχεια, επιλέγουμε τη δειγματοληπτική διαδικασία. Ένα δείγμα είναι ένα υποσύνολο ενός μεγαλύτερου πληθυσμού και πρέπει να απαντηθούν διάφορες ερωτήσεις πριν επιλέξουμε το δείγμα. Το δείγμα, λοιπόν, πρέπει να περιλαμβάνει όλους εκείνους τους ανθρώπους, των οποίων οι απόψεις, συμπεριφορές, προτιμήσεις, νοοτροπίες κλπ. θα βοηθήσουν στη λήψη απόφασης του marketer. Το επόμενο στάδιο είναι η συλλογή των δεδομένων. Η διαδικασία του Data Collection γίνεται από το τμήμα Field, το οποίο είναι αρμόδιο εκτός των άλλων για την εξεύρεση και τη συνέντευξη και ενημέρωση των ερευνητών. Μετά τη συλλογή των δεδομένων, είναι η σειρά της επεξεργασίας και ανάλυσης αυτών. Ο σκοπός αυτής της ανάλυσης είναι να ερμηνεύουμε και να σχεδιάσουμε συμπεράσματα από το σύνολο των συλλεχθέντων στοιχείων. Αφού αποπερατωθεί η ανάλυση των δεδομένων, ο marketer πρέπει να προετοιμάσει το report και να μεταφέρει τα αποτελέσματα και τις προτάσεις του (points for action) στους executives. Τέλος, το τμήμα Marketing πρέπει να αποφασίσει εάν οι προτάσεις έγιναν πράξη και γιατί ή γιατί δεν έγιναν. / -

O efeito de instalação de estacas cravadas em areia, que promove um aumento da compacidade do solo e de seus parâmetros de resistência e deformabilidade, é analisado nesta dissertação em conjunto com o efeito de grupo. O procedimento estabelecido objetiva quantificar a melhoria após a instalação de um grupo de estacas e foi desenvolvido nesta pesquisa a partir da contribuição de Alves (1998). Para a utilização deste procedimento, foram propostas correlações do módulo de Young do solo em função da resistência à penetração NSPT (ou da resistência à penetração normalizada N60), a partir de um banco de dados obtido da literatura, visando não apenas as aplicações implementadas neste estudo, como também nortear futuras aplicações práticas. Na análise de casos documentados da literatura, foi utilizado o programa Plaxis 3D Foundation. Inicialmente foi procedida a calibração do programa através da reprodução de um caso de instrumentação de um grupo de estacas na argila rija de Londres, muito bem documentado por Cooke et al. (1980). Os resultados das simulações, com parâmetros do solo obtidos da literatura, se ajustaram muito bem aos resultados experimentais, seja nos valores de recalque, seja nos diagramas de transferência de carga. Posteriormente, a melhoria do solo por efeito da instalação foi estimada, através do procedimento estabelecido nesta dissertação, baseado em Alves (1998), a partir das características iniciais do solo antes da instalação de um grupo de estacas em modelo reduzido em solo arenoso. Obtidos os parâmetros melhorados previstos para o solo após a instalação, estes foram aplicados a um grupo de 9 estacas, para diferentes espaçamentos relativos, como dados de entrada no programa Plaxis. A comparação dos resultados experimentais com aqueles simulados numericamente sinaliza para o potencial tanto do programa Plaxis, na reprodução do comportamento do conjunto, como do modelo de Alves (1998), que norteou o procedimento proposto nesta pesquisa. O fator de escala, ressaltado na dissertação, bem como outros aspectos relevantes observados nas análises realizadas, são propostos como temas futuros a serem investigados na continuidade desta linha de pesquisa. / The installation effect of piles driven in sand, which increases soill density, strength and deformability parameters, is analyzed in this research, together with the pile-group effect. The procedure has been established aiming at the soil improvement achieved after installation of the piles. The proposed procedure is based on Alves (1998). In order to apply the procedure, correlations between the soil Young modulus and the NSPT blow count (or the normalized N60) have been established from a database obtained in the literature, aiming not only at the applications implemented in this study but also as a guideline for future practical applications. The program Plaxis 3D Foundation has been used in the analysis of documented cases from the literature, as an initial step to implement the procedure proposed in this thesis. Initially, the program has been calibrated by application to a well-documented piling group in the London stiff clay by Cooke et al. (1980). The results of the numerical simulations with soil parameters obtained from the literature have been very close to the experimental values, not only for the settlements, but also for the load transfer diagrams. The soil improvement due to pile installation has been implemented in this thesis based on Alves (1998). It has been applied to a case of small-scale model piling group in sands, taking into consideration the initial characteristics of the soil before installation and the soil improvement after driving. The updated parameters, together with different pile spacing in the 9-pile group, have been entered as input in the program Plaxis 3D Foundation. The comparison of the experimental results with those numerically simulated shows the potentiality of Plaxis program, due to its ability to properly reproduce the piling group behavior, and also of the model by Alves (1998), which guided the soil improvement procedure proposed in this research. The scale factor, highlighted in the thesis, as well as other relevant factors observed in the analyses, are topics for further investigation in continuity of this research.

Conselho Nacional de Desenvolvimento Científico e Tecnológico / Castor bean is an oil and biodiesel source. Besides, it has a high seed yield potential in the State of Rio Grande do Sul. The objective of this work was to establish the relationship among the variables: seeds, seedlings, adult plants, grain yield and oil quality in two castor bean hybrids. The experiment was conducted at the Federal University of Santa Maria during the 2007/2008 growing season using the hybrids Sara and Lyra. Three plots of each hybrid were sown; each plot was formed by 13 rows with different distances between plants: 0.4, 0.6 and 0.8 m. Before, an evaluation was conducted in regard to individual characteristics of the seeds on a total of 90 seeds for each hybrids. In each plot, the two central rows consisted of 30 plants and these were evaluated from emergence until harvest comprising seedling, adult plant, grain yield and oil quality. In each group of variables the following propositions were tested: multivariate normality, deviation homocedasticity, and linearity in order to analyze the canonical correlations among groups. For the Sara hybrid, the relationships among groups were established mainly by: seeds of higher weight, width and length positively affected seedling vigor; seedlings correlated with longer epicotyls 14 days after emergence (DAE) and longer hipocotyl 7 DAE that were associated with adult plants height at beginning of flowering, average leaf area, average raceme length, average number of capsules per raceme and negatively the raceme average length; higher adult plants at flowering had more capsules per raceme, higher grain yield and lower average raceme length. In regard to the hybrid Lyra, weight and size of seeds did not affect seedling vigor. The relationship among groups was established by: seedlings with longest epicotyls 7 DAE, smallest epicotyls 14 DAE, less degree-days to emergence and greater number of days to emit the first true leaves determined higher number of inflorescences and average leaf area and lower plant height at flowering and degree-days to flowering; adult plants with the higher number of inflorescences positively affected raceme number. Plants of the hybrid Sara presented significant canonical correlations with most of the variables except for oil quality. For the hybrid Lyra, significant canonical correlations were observed only between the variables of seedlings and adult plants; and, between variables of adult plants and grain yield. For the hybrids studied, the oil quality is not determined by the productive variables. / A mamona é uma matéria-prima que possui excelente qualidade de óleo para a indústria ricinoquímica e para a produção de biodiesel. Além disso, apresenta elevado potencial produtivo no estado do Rio Grande do Sul. O objetivo desse trabalho foi identificar as relações entre as variáveis de: sementes, plântulas, plantas adultas, produção de grãos e qualidade de óleo, em dois híbridos de mamona. O experimento com a cultura de mamoneira, híbridos Sara e Lyra foi conduzido no ano agrícola 2007/2008, no Departamento de Fitotecnia da Universidade Federal de Santa Maria. Foram semeadas três parcelas do híbrido Sara e três do híbrido Lyra. Cada parcela foi constituída por 13 fileiras, com espaçamento entre plantas de 0,4; 0,6 e 0,8 m. Anterior a instalação do experimento em campo, foram realizadas avaliações individuais das características físicas das sementes (180 sementes), sendo 90 do híbrido Lyra e 90 do híbrido Sara. Em cada parcela, as duas linhas centrais foram ocupadas com 30 sementes e as plantas originadas foram avaliadas desde a emergência até a colheita. Além das variáveis de sementes, foram avaliadas as variáveis de: plântulas, plantas adultas, produção de grãos e óleo. Em cada grupo de variáveis foram testadas as pressuposições: normalidade multivariada, homocedasticidade dos desvios, multicolinearidade e linearidade, para a análise de correlação canônica entre os grupos. No híbrido Sara, as relações entre os grupos são estabelecidas principalmente por: sementes de maior massa, largura e comprimento, influenciam positivamente o vigor de plântulas; plântulas com maior comprimento do epicótilo aos 14 DAE (Dias após a emergência) e com maior comprimento do hipocótilo aos sete DAE, influenciam positivamente a altura da planta no início do florescimento, a área foliar média - método do comprimento da nervura principal, o número de inflorescências, o número médio de cápsulas por rácemo e, negativamente, o comprimento médio dos rácemos; plantas adultas com maior número de inflorescências influenciam positivamente o número de rácemos por planta e, plantas adultas mais altas no florescimento determinam maior número médio de cápsulas por rácemo, maior rendimento e menor comprimento médio dos rácemos. No híbrido Lyra, sementes de maior massa e tamanho não influenciam o vigor de plântulas. As relações entre os grupos são estabelecidas principalmente por: plântulas com maior comprimento do hipocótilo aos sete DAE, menor comprimento do epicótilo aos 14 DAE, menos grausdia para emergência e maior quantidade de dias para emitir as primeiras folhas verdadeiras determinam maior número de inflorescências e área foliar média - método do comprimento da nervura principal e, com menor altura da planta no florescimento e graus-dia para o início do florescimento; plantas adultas com maior número de inflorescências influenciam positivamente o número de rácemos. Plantas do híbrido Sara apresentam correlações canônicas significativas entre a maioria dos grupos de variáveis, exceto para a qualidade de óleo. No híbrido Lyra, as correlações canônicas significativas ocorrem somente entre os grupos de plântulas e plantas adultas; e, entre plantas adultas e produção. A qualidade do óleo de mamona, para os híbridos Sara e Lyra, não é determinada pelas variáveis de produção.

Made available in DSpace on 2014-10-09T12:43:31Z (GMT). No. of bitstreams: 0 / Made available in DSpace on 2014-10-09T13:56:05Z (GMT). No. of bitstreams: 1 06635.pdf: 4560564 bytes, checksum: 677cab6b14a66c0989ef5859a83679e3 (MD5) / Dissertacao (Mestrado) / IPEN/D / Instituto de Pesquisas Energeticas e Nucleares - IPEN/CNEN-SP

This dissertation studies the cost of carry relationship and the international dynamics of mispricing, price and volatility in the three Nikkei futures markets - the Osaka Exchange (OSE), the Singapore Exchange (SGX) and the Chicago Mercantile Exchange (CME). Previous research does not fully consider the unique characteristics of the triple-listed Nikkei futures contracts, or the price and volatility dynamics in the three Nikkei futures exchanges at the same time. This dissertation makes a significant contribution to the existing literature. In particular, with a comprehensive new 19-year sample period, this dissertation helps deepen the understanding of the Nikkei spot-futures equilibrium and arbitrage behaviour, cross-border information transmission mechanism, and futures market integration. The first topic of the dissertation is to study the cost of carry relationship, mispricing and index arbitrage in the three Nikkei markets. The standard cost of carry model is adjusted for each Nikkei futures contract by allowing for the triple-listing nature and key institutional differences. Based on this, the economic significance of the Nikkei mispricing is explored in the presence of transaction costs. The static behaviour of the mispricing suggests that it is difficult especially for institutional investors to make arbitrage profits in the OSE and SGX, and that index arbitrage in the CME is not strictly risk-free due to the exchange rate effect. Smooth transition models are used to study the dynamic behaviour of the mispricing in the three markets. The results show that mean reversion in mispricing and limits to arbitrage are driven more by transaction costs than by heterogeneous arbitrageurs in the Nikkei markets. The second topic of the dissertation is to investigate the price discovery process in individual Nikkei markets and across the Nikkei futures markets. With smooth transition error correction models, this dissertation reports the leading role of the futures prices in the pre-crisis period and the leading role of the spot prices in the post-crisis period, in the first-moment information transmission process. Moreover, there is evidence of asymmetric adjustments in the Nikkei prices and volatilities. The cross-border dynamics suggest that the foreign Nikkei markets (the CME and SGX) act as the main price discovery vehicle, which implies the key functions of the equivalent, offshore markets in futures market globalisation. The third topic of the dissertation is to study the volatility transmission process in individual Nikkei markets and across the Nikkei futures markets, from the perspectives of the volatility interactions in and across the Nikkei markets and of the dynamic Nikkei market linkages. This dissertation finds bidirectional volatility spillover effects between the Nikkei spot and futures markets, and the information leadership of the foreign Nikkei markets (the CME and SGX) in the second-moment information transmission process across the border. It further examines the dynamic conditional correlations between the Nikkei markets. The results point to a dramatic integration process with strongly persistent and stable Nikkei market co-movements over time.

La forte augmentation du volume de données multimédia impose la mise au point de solutions adaptées pour une exploration visuelle efficace des bases multimédia. Après avoir examiné les processus de visualisation mis en jeu, nous remarquons que ceci demande une structuration des données. L’objectif principal de cette thèse est de proposer et d’étudier ces méthodes de structuration des bases multimédia en vue de leur exploration visuelle.Nous commençons par un état de l’art détaillant les données et les mesures que nous pouvons produire en fonction de la nature des variables décrivant les données. Suit un examen des techniques de structuration par projection et classification. Nous présentons aussi en détail la technique du Clustering Spectral sur laquelle nous nous focaliserons ensuite.Notre première réalisation est une méthode originale de production et fusion de métriques par corrélation de rang. Nous testons cette première méthode sur une base multimédia issue de la vidéothèque d’un festival de films. Nous continuons ensuite par la mise au point d’une méthode de classification supervisée par corrélation que nous testons avec les données vidéos d’un challenge de la communauté multimédia. Ensuite nous nous focalisons sur les techniques du Clustering Spectral. Nous testons une technique de Clustering Spectral supervisée que nous comparons aux techniques de l’état de l’art. Et pour finir nous examinons des techniques du Clustering Spectral semi-supervisé actif. Dans ce contexte, nous proposons et validons des techniques de propagation d’annotations et des stratégies permettant d’améliorer la convergence de ces méthodes de classement. / The large increase in multimedia data volume requires the development of effective solutions for visual exploration of multimedia databases. After reviewing the visualization process involved, we emphasis the need of data structuration. The main objective of this thesis is to propose and study clustering and classification of multimedia database for their visual exploration.We begin with a state of the art detailing the data and the metrics we can produce according to the nature of the variables describing each document. Follows a review of the projection and classification techniques. We also present in detail the Spectral Clustering method.Our first contribution is an original method that produces fusion of metrics using rank correlations. We validate this method on an animation movie database coming from an international festival. Then we propose a supervised classification method based on rank correlation. This contribution is evaluated on a multimedia challenge dataset. Then we focus on Spectral Clustering methods. We test a supervised Spectral Clustering technique and compare to state of the art methods. Finally we examine active semi-supervised Spectral Clustering methods. In this context, we propose and validate constraint propagation techniques and strategies to improve the convergence of these active methods.