The fact that clustering is perhaps the most used technique for exploratory data analysis is only a semaphore that underlines its fundamental importance. The general problem statement that broadly describes clustering as the identification and classification of patterns into coherent groups also implicitly indicates it's utility in other tasks such as supervised learning. In the past decade and a half there have been two developments that have altered the landscape of research in clustering: One is improved results by the increased use of graph theoretic techniques such as spectral clustering and the other is the study of clustering with respect to its relevance in semi-supervised learning i.e. using unlabeled data for improving prediction accuracies. In this work an attempt is made to make contributions to both these aspects. Thus our contributions are two-fold: First, we identify some general issues with the spectral clustering framework and while working towards a solution, we introduce a new algorithm which we call "Regularity Clustering" which makes an attempt to harness the power of the Szemeredi Regularity Lemma, a remarkable result from extremal graph theory for the task of clustering. Secondly, we investigate some practical and useful strategies for using clustering unlabeled data in boosting prediction accuracy. For all of these contributions we evaluate our methods against existing ones and also apply these ideas in a number of settings.

Cette thèse porte sur la façon dont la mobilisation des personnes en situation d'exclusion peut contribuer à la fabrication des politiques publiques. L'observation empirique prend appui sur une ONG brésilienne, travaillant avec des enfants et adolescents de deux favelas de Recife et qui utilise la notion de citoyenneté comme axe central de son projet politico-pédagogique. Nous faisons l'hypothèse que la notion recouvre un ensemble d'attentes espérées par le Brésil dans sa phase de redémocratisation et mises en échec par les politiques néolibérales. Après une première partie s'intéressant à la difficile mise en place des politiques publiques au Brésil et une deuxième qui tente une reconstruction de la notion de citoyenneté, la troisième partie s'intéresse au cadre de sa mise en œuvre. Associée à la démocratie, la citoyenneté devient un « art du vivre ensemble » supposant la reconnaissance d'une communauté politique qui ouvre à l'expérience du « participable » et du « partageable ». L'espace public est dès lors l'élément central de cette mise en scène autorisant l'existence de communautés particulières et permettant de répondre à la fois aux besoins d'assignation des individus et de diversité culturelle. Les notions d'égalité et de liberté complètent les notions qui précèdent en élargissant le champ de compréhension de la façon dont la citoyenneté peut devenir effective. La conclusion pointe la nécessaire formation du « sujet-citoyen » que le Groupe Pé no Chão nous a permis de concevoir et place la construction des identités individuelles et collectives comme élément de transformation sociale. / This thesis discusses the ways in which the motivating forces of marginalized people may contribute towards the development of public policy. The empirical observation uses a Brazilian NGO developing “Social Education in the Street”, which works with children and teenagers in two favelas in Recife. It uses the notion of citizenship as the central core of its politico-educational project. Observation demonstrates that this NGO is not the only one in Brazil to use the term of citizenship. The term is used by numerous players in civil and political society, and is almost excessively used. We hypothesise that the term covers a whole range of Brazil's expectations in its phase of re-democratisation, which are frustrated by neo-liberal policies. The first part of the research investigates the difficulties of implementing public policies on Brazil. The second part, a reconstruction of the notion of citizenship. Using this interpretation of citizenship, the third part investigates its implementation. Associated to democracy, citizenship becomes “the art of living together”, presupposing the recognition of a political community which allows the experience of taking part and sharing. Henceforth public space is the central element of this scenario, authorising the existence of individual communities and enabling the fulfilment of needs both of belonging and of cultural diversity. These notions are complemented by those of equality and liberty, widening understanding of the ways in which the citizen may become effective. The conclusion highlights the training necessary for the “subject-citizen” that the Pé no Chão Group enabled us to develop and positions the construction of individual and collective identities as an element of social transformation.

Digitized by Kansas Correctional Industries

Includes bibliographical references (leaves 19-45). / Digitized by Kansas Correctional Industries

Title from accompanying document. / Digitized by Kansas Correctional Industries

An ensemble of classifiers succeeds in improving the accuracy of the whole when thecomponent classifiers are both diverse and accurate. Diversity is required to ensure that theclassifiers make uncorrelated errors. Theoretical and experimental approaches from previousresearch show very low correlation between ensemble accuracy and diversity measure.Introducing Proposed Compound diversity functions by Albert Hung-Ren KO and RobertSabourin, (2009), by combining diversities and performances of individual classifiers exhibitstrong correlations between the diversities and accuracy. To be consistent with existingarguments compound diversity of measures are evaluated and compared with traditionaldiversity measures on different problems. Evaluating diversity of errors and comparison withmeasures are significant in this study. The results show that compound diversity measuresare better than ordinary diversity measures. However, the results further explain evaluation ofdiversity of errors on available data. / Program: Magisterutbildning i informatik

Algoritmos de aprendizado de máquina são frequentemente os mais indicados em uma grande variedade de aplicações de mineração dados. Entretanto, a maioria das pesquisas em aprendizado de máquina refere-se ao problema bem definido de encontrar um modelo (geralmente de classificação) de um conjunto de dados pequeno, relativamente bem preparado para o aprendizado, no formato atributo-valor, no qual os atributos foram previamente selecionados para facilitar o aprendizado. Além disso, o objetivo a ser alcançado é simples e bem definido (modelos de classificação precisos, no caso de problemas de classificação). Mineração de dados propicia novas direções para pesquisas em aprendizado de máquina e impõe novas necessidades para outras. Com a mineração de dados, algoritmos de aprendizado estão quebrando as restrições descritas anteriormente. Dessa maneira, a grande contribuição da área de aprendizado de máquina para a mineração de dados é retribuída pelo efeito inovador que a mineração de dados provoca em aprendizado de máquina. Nesta tese, exploramos alguns desses problemas que surgiram (ou reaparecem) com o uso de algoritmos de aprendizado de máquina para mineração de dados. Mais especificamente, nos concentramos seguintes problemas: Novas abordagens para a geração de regras. Dentro dessa categoria, propomos dois novos métodos para o aprendizado de regras. No primeiro, propomos um novo método para gerar regras de exceção a partir de regras gerais. No segundo, propomos um algoritmo para a seleção de regras denominado Roccer. Esse algoritmo é baseado na análise ROC. Regras provêm de um grande conjunto externo de regras e o algoritmo proposto seleciona regras baseado na região convexa do gráfico ROC. Proporção de exemplos entre as classes. Investigamos vários aspectos relacionados a esse tópico. Primeiramente, realizamos uma série de experimentos em conjuntos de dados artificiais com o objetivo de testar nossa hipótese de que o grau de sobreposição entre as classes é um fator complicante em conjuntos de dados muito desbalanceados. Também executamos uma extensa análise experimental com vários métodos (alguns deles propostos neste trabalho) para balancear artificialmente conjuntos de dados desbalanceados. Finalmente, investigamos o relacionamento entre classes desbalanceadas e pequenos disjuntos, e a influência da proporção de classes no processo de rotulação de exemplos no algoritmo de aprendizado de máquina semi-supervisionado Co-training. Novo método para a combinação de rankings. Propomos um novo método, chamado BordaRank, para construir ensembles de rankings baseado no método de votação borda count. BordaRank pode ser aplicado em qualquer problema de ordenação binária no qual vários rankings estejam disponíveis. Resultados experimentais mostram uma melhora no desempenho com relação aos rankings individuais, alem de um desempenho comparável com algoritmos mais sofisticados que utilizam a predição numérica, e não rankings, para a criação de ensembles para o problema de ordenação binária. / Machine learning algorithms are often the most appropriate algorithms for a great variety of data mining applications. However, most machine learning research to date has mainly dealt with the well-circumscribed problem of finding a model (generally a classifier) given a single, small and relatively clean dataset in the attribute-value form, where the attributes have previously been chosen to facilitate learning. Furthermore, the end-goal is simple and well-defined, such as accurate classifiers in the classification problem. Data mining opens up new directions for machine learning research, and lends new urgency to others. With data mining, machine learning is now removing each one of these constraints. Therefore, machine learning's many valuable contributions to data mining are reciprocated by the latter's invigorating effect on it. In this thesis, we explore this interaction by proposing new solutions to some problems due to the application of machine learning algorithms to data mining applications. More specifically, we contribute to the following problems. New approaches to rule learning. In this category, we propose two new methods for rule learning. In the first one, we propose a new method for finding exceptions to general rules. The second one is a rule selection algorithm based on the ROC graph. Rules come from an external larger set of rules and the algorithm performs a selection step based on the current convex hull in the ROC graph. Proportion of examples among classes. We investigated several aspects related to this issue. Firstly, we carried out a series of experiments on artificial data sets in order to verify our hypothesis that overlapping among classes is a complicating factor in highly skewed data sets. We also carried out a broadly experimental analysis with several methods (some of them proposed by us) that artificially balance skewed datasets. Our experiments show that, in general, over-sampling methods perform better than under-sampling methods. Finally, we investigated the relationship between class imbalance and small disjuncts, as well as the influence of the proportion of examples among classes in the process of labelling unlabelled cases in the semi-supervised learning algorithm Co-training. New method for combining rankings. We propose a new method called BordaRanking to construct ensembles of rankings based on borda count voting, which could be applied whenever only the rankings are available. Results show an improvement upon the base-rankings constructed by taking into account the ordering given by classifiers which output continuous-valued scores, as well as a comparable performance with the fusion of such scores.

This thesis introduces novel ideas in subsurface reservoir model calibration known as History Matching in the reservoir engineering community. The target of history matching is to mimic historical pressure and production data from the producing wells with the output from the reservoir simulator for the sole purpose of reducing uncertainty from such models and improving confidence in production forecast. Ensemble based methods such as the Ensemble Kalman Filter (EnKF) and Ensemble Smoother with Multiple Data Assimilation (ES-MDA) as been proposed for history matching in literature. EnKF/ES-MDA is a Monte Carlo ensemble nature filter where the representation of the covariance is located at the mean of the ensemble of the distribution instead of the uncertain true model. In EnKF/ES-MDA calculation of the gradients is not required, and the mean of the ensemble of the realisations provides the best estimates with the ensemble on its own estimating the probability density. However, because of the inherent assumptions of linearity and Gaussianity of petrophysical properties distribution, EnKF/ES-MDA does not provide an acceptable history-match and characterisation of uncertainty when tasked with calibrating reservoir models with channel like structures. One of the novel methods introduced in this thesis combines a successive parameter and shape reconstruction using level set functions (EnKF/ES-MDA-level set) where the spatial permeability fields' indicator functions are transformed into signed distances. These signed distances functions (better suited to the Gaussian requirement of EnKF/ES-MDA) are then updated during the EnKF/ES-MDA inversion. The method outperforms standard EnKF/ES-MDA in retaining geological realism of channels during and after history matching and also yielded lower Root-Mean-Square function (RMS) as compared to the standard EnKF/ES-MDA. To improve on the petrophysical reconstruction attained with the EnKF/ES-MDA-level set technique, a novel parametrisation incorporating an unsupervised machine learning method for the recovery of the permeability and porosity field is developed. The permeability and porosity fields are posed as a sparse field recovery problem and a novel SELE (Sparsity-Ensemble optimization-Level-set Ensemble optimisation) approach is proposed for the history matching. In SELE some realisations are learned using the K-means clustering Singular Value Decomposition (K-SVD) to generate an overcomplete codebook or dictionary. This dictionary is combined with Orthogonal Matching Pursuit (OMP) to ease the ill-posed nature of the production data inversion, converting our permeability/porosity field into a sparse domain. SELE enforces prior structural information on the model during the history matching and reduces the computational complexity of the Kalman gain matrix, leading to faster attainment of the minimum of the cost function value. From the results shown in the thesis; SELE outperforms conventional EnKF/ES-MDA in matching the historical production data, evident in the lower RMS value and a high geological realism/similarity to the true reservoir model.

Nesta tese são apresentados estudos do impacto de armadilhas no desempenho elétrico de MOSFETs em nível de circuito e um simulador Ensamble Monte Carlo (EMC) é apresentado visando a análise do impacto de armadilhas em nível de dispositivo. O impacto de eventos de captura e emissão de portadores por armadilhas na performance e confiabilidade de circuitos é estudada. Para tanto, um simulador baseado em SPICE que leva em consideração a atividade de armadilhas em simulações transientes foi desenvolvido e é apresentado seguido de estudos de caso em células SRAM, circuitos combinacionais, ferramentas de SSTA e em osciladores em anel. Foi também desenvolvida uma ferramenta de simulação de dispositivo (TCAD) atomística baseada no método EMC para MOSFETs do tipo p. Este simulador é apresentado em detalhes e seu funcionamento é testado conceitualmente e através de comparações com ferramentas comerciais similares. / This thesis presents studies on the impact of charge traps in MOSFETs at the circuit level, and a Ensemble Monte Carlo (EMC) simulation tool is developed to perform analysis on trap impact on PMOSFETs. The impact of charge trapping on the performance and reliability of circuits is studied. A SPICE based simulator, which takes into account the trap activity in transient simulations, was developed and used on case studies of SRAM, combinational circuits, SSTA tools and ring oscillators. An atomistic device simulator (TCAD) for modeling of p-type MOSFETs based on the EMC simulation method was also developed. The simulator is explained in details and its well function is tested.

A monoaminoxidase (MAO) é uma enzima importante, que pode atuar como alvo terapêutico. Inibidores da MAO-A apresentam atividade no tratamento de distúrbios de humor, enquanto os inibidores seletivos da MAO-B tem uso, especialmente, no tratamento da Doença de Parkinson. O conhecimento das interações ENZIMA-INIBIDOR é importante no planejamento de fármacos. Nesse contexto, foram realizados estudos das enzimas MAO-A e MAO-B com diferentes ligantes, através da combinação das metodologias de docking, Dinâmica Molecular e Ensemble Docking. Foram escolhidos os ligantes derivados da 1,4-naftoquinona (1,4-NQ), lapachol, menadiona, norlapachol, A2, B2 e C2, os inibidores comerciais clorgilina (MAO-A) e selegilina (MAO-B) e os substratos naturais serotonina (MAO-A) e dopamina (MAO-B). Os resultados do docking mostraram interação de todos os ligantes com algum dos resíduos da "gaiola aromática" (FAD, Tyr407/Tyr444 para MAO-A, Tyr398/Tyr435 para MAO-B), uma importante região catalítica da MAO. Além disso, a seletividade observada experimentalmente da menadiona com a MAO-B também foi observada no docking. Através da DM, foi possível observar algumas diferenças conformacionais entre as estruturas da MAO-A e MAO-B, que podem explicar a seletividade entre as duas isoformas, como por exemplo, distâncias entre resíduos do sítio ativo e ligações de hidrogênio. A partir do Ensemble Docking, foi verificado que a conformação do receptor influencia significativamente o escore das interações ENZIMA+LIGANTE para ligantes volumosos. / Monoamine oxidase (MAO) is an important enzyme that acts as therapeutic target. MAO-A inhibitors show pharmacological activity in the treatment of mood disorders, whereas MAO-B inhibitors are used especially in treatment of Parkinson's Disease. Knowledge of enzyme-inhibitor interactions is important in drug design. Therefore, studies of MAO-A and MAO-B enzymes with different ligands were performed by combining docking, Molecular Dynamics and Ensemble Docking methodologies. Ligands derived from 1,4-naphthoquinone, lapachol, menadione, nor-lapachol, A2, B2, C2, commercial inhibitors clorgyline (MAO-A) and selegiline (MAO-B) and the natural substrates serotonin (MAO-A) and dopamine (MAO-B) were chosen. The docking results shows interactions of all ligands with some residue of the "aromatic cage” (FAD cofactor, Tyr407/Tyr444 for MAO-A and Tyr398/Tyr435 for MAO-B), an important catalytic region of MAO. Furthermore, the experimentally observed selectivity of menadione with MAO-B was also observed by Docking. In Molecular Dynamics results, conformational differences were observed between MAO-A and MAO-B structures, which could explain the selectivity observed between isoforms, e.g. distances between residues of the active site and hydrogen bonds. Ensemble Docking results shows that the conformation of the receptor significantly influence the score of ENZYME+LIGAND interactions for bulky ligands.