Casamento de padrões de pontos em Espaços Euclidianos é um dos problemas fundamentais em reconhecimento de padrões, tendo aplicações que vão desde Visão Computacional até Química Computacional. Sempre que dois padrões complexos estão codi- ficados em termos de dois conjuntos de pontos que identificam suas características fundamentais, sua comparação pode ser vista como um problema de casamento de padrões de pontos. Este trabalho propõe uma abordagem unificada para os problemas de casamento exato e inexato de padrões de pontos em Espaços Euclidianos de dimensão arbitrária. No caso de casamento exato, é garantida a obtenção de uma solução ótima. Para casamento inexato (quando ruído está presente), resultados experimentais confirmam a validade da abordagem. Inicialmente, considera-se o problema de casamento de padrões de pontos como um problema de casamento de grafos ponderados. O problema de casamento de grafos ponderados é então formulado como um problema de inferência Bayesiana em um modelo gráfico probabilístico. Ao explorar certos vínculos fundamentais existentes em padrões de pontos imersos em Espaços Euclidianos, provamos que, para o casamento exato de padrões de pontos, um modelo gráfico simples é equivalente ao modelo completo. É possível mostrar que inferência probabilística exata neste modelo simples tem complexidade polinomial para qualquer dimensionalidade do Espaço Euclidiano em consideração. Experimentos computacionais comparando esta técnica com a bem conhecida baseada em relaxamento probabilístico evidenciam uma melhora significativa de desempenho para casamento inexato de padrões de pontos. A abordagem proposta é signi- ficativamente mais robusta diante do aumento do tamanho dos padrões envolvidos. Na ausência de ruído, os resultados são sempre perfeitos. / Point pattern matching in Euclidean Spaces is one of the fundamental problems in Pattern Recognition, having applications ranging from Computer Vision to Computational Chemistry. Whenever two complex patterns are encoded by two sets of points identifying their key features, their comparison can be seen as a point pattern matching problem. This work proposes a single approach to both exact and inexact point set matching in Euclidean Spaces of arbitrary dimension. In the case of exact matching, it is assured to find an optimal solution. For inexact matching (when noise is involved), experimental results confirm the validity of the approach. We start by regarding point pattern matching as a weighted graph matching problem. We then formulate the weighted graph matching problem as one of Bayesian inference in a probabilistic graphical model. By exploiting the existence of fundamental constraints in patterns embedded in Euclidean Spaces, we prove that for exact point set matching a simple graphical model is equivalent to the full model. It is possible to show that exact probabilistic inference in this simple model has polynomial time complexity with respect to the number of elements in the patterns to be matched. This gives rise to a technique that for exact matching provably finds a global optimum in polynomial time for any dimensionality of the underlying Euclidean Space. Computational experiments comparing this technique with well-known probabilistic relaxation labeling show significant performance improvement for inexact matching. The proposed approach is significantly more robust under augmentation of the sizes of the involved patterns. In the absence of noise, the results are always perfect.

Casamento de padrões de pontos em Espaços Euclidianos é um dos problemas fundamentais em reconhecimento de padrões, tendo aplicações que vão desde Visão Computacional até Química Computacional. Sempre que dois padrões complexos estão codi- ficados em termos de dois conjuntos de pontos que identificam suas características fundamentais, sua comparação pode ser vista como um problema de casamento de padrões de pontos. Este trabalho propõe uma abordagem unificada para os problemas de casamento exato e inexato de padrões de pontos em Espaços Euclidianos de dimensão arbitrária. No caso de casamento exato, é garantida a obtenção de uma solução ótima. Para casamento inexato (quando ruído está presente), resultados experimentais confirmam a validade da abordagem. Inicialmente, considera-se o problema de casamento de padrões de pontos como um problema de casamento de grafos ponderados. O problema de casamento de grafos ponderados é então formulado como um problema de inferência Bayesiana em um modelo gráfico probabilístico. Ao explorar certos vínculos fundamentais existentes em padrões de pontos imersos em Espaços Euclidianos, provamos que, para o casamento exato de padrões de pontos, um modelo gráfico simples é equivalente ao modelo completo. É possível mostrar que inferência probabilística exata neste modelo simples tem complexidade polinomial para qualquer dimensionalidade do Espaço Euclidiano em consideração. Experimentos computacionais comparando esta técnica com a bem conhecida baseada em relaxamento probabilístico evidenciam uma melhora significativa de desempenho para casamento inexato de padrões de pontos. A abordagem proposta é signi- ficativamente mais robusta diante do aumento do tamanho dos padrões envolvidos. Na ausência de ruído, os resultados são sempre perfeitos. / Point pattern matching in Euclidean Spaces is one of the fundamental problems in Pattern Recognition, having applications ranging from Computer Vision to Computational Chemistry. Whenever two complex patterns are encoded by two sets of points identifying their key features, their comparison can be seen as a point pattern matching problem. This work proposes a single approach to both exact and inexact point set matching in Euclidean Spaces of arbitrary dimension. In the case of exact matching, it is assured to find an optimal solution. For inexact matching (when noise is involved), experimental results confirm the validity of the approach. We start by regarding point pattern matching as a weighted graph matching problem. We then formulate the weighted graph matching problem as one of Bayesian inference in a probabilistic graphical model. By exploiting the existence of fundamental constraints in patterns embedded in Euclidean Spaces, we prove that for exact point set matching a simple graphical model is equivalent to the full model. It is possible to show that exact probabilistic inference in this simple model has polynomial time complexity with respect to the number of elements in the patterns to be matched. This gives rise to a technique that for exact matching provably finds a global optimum in polynomial time for any dimensionality of the underlying Euclidean Space. Computational experiments comparing this technique with well-known probabilistic relaxation labeling show significant performance improvement for inexact matching. The proposed approach is significantly more robust under augmentation of the sizes of the involved patterns. In the absence of noise, the results are always perfect.

La capacité des robots mobiles à se localiser précisément par rapport à leur environnement est indispensable à leur autonomie. Pour ce faire, les robots exploitent les données acquises par des capteurs qui observent leur état interne, tels que centrales inertielles ou l’odométrie, et les données acquises par des capteurs qui observent l’environnement, telles que les caméras et les Lidars. L’exploitation de ces derniers capteurs a suscité le développement de solutions qui estiment conjointement la position du robot et la position des éléments dans l'environnement, appelées SLAM (Simultaneous Localization and Mapping). Pour gérer le bruit des données provenant des capteurs, les solutions pour le SLAM sont mises en œuvre dans un contexte probabiliste. Les premiers développements étaient basés sur le filtre de Kalman étendu, mais des développements plus récents utilisent des modèles graphiques probabilistes pour modéliser le problème d’estimation et de le résoudre grâce à techniques d’optimisation. Cette thèse exploite cette dernière approche et propose deux techniques distinctes pour les véhicules terrestres autonomes: une utilisant la vision monoculaire, l’autre un Lidar. L’absence d’information de profondeur dans les images obtenues par une caméra a mené à l’utilisation de paramétrisations spécifiques pour les points de repères qui isolent la profondeur inconnue dans une variable, concentrant la grande incertitude sur la profondeur dans un seul paramètre. Une de ces paramétrisations, nommé paramétrisation pour l’angle de parallaxe (ou PAP, Parallax Angle Parametrization), a été introduite dans le contexte du problème d’ajustement de faisceaux, qui traite l’ensemble des données en une seule étape d’optimisation globale. Nous présentons comment exploiter cette paramétrisation dans une approche incrémentale de SLAM à base de modèles graphiques, qui intègre également les mesures de mouvement du robot. Les Lidars peuvent être utilisés pour construire des solutions d’odométrie grâce à un recalage séquentiel des nuages de points acquis le long de la trajectoire. Nous définissons une couche basée sur les modèles graphiques au dessus d’une telle couche d’odométrie, qui utilise l’algorithme ICP (Iterative Closest Points). Des repères clefs (keyframes) sont définis le long de la trajectoire du robot, et les résultats de l’algorithme ICP sont utilisés pour construire un graphe de poses, exploité pour résoudre un problème d’optimisation qui permet la correction de l’ensemble de la trajectoire du robot et de la carte de l’environnement à suite des fermetures de boucle.Après une introduction à la théorie des modèles graphiques appliquée au problème de SLAM, le manuscrit présente ces deux approches. Des résultats simulés et expérimentaux illustrent les développements tout au long du manuscrit, en utilisant des jeux des données classiques et obtenus au laboratoire. / A strong requirement to deploy autonomous mobile robots is their capacity to localize themselves with a certain precision in relation to their environment. Localization exploits data gathered by sensors that either observe the inner states of the robot, like acceleration and speed, or the environment, like cameras and Light Detection And Ranging (LIDAR) sensors. The use of environment sensors has triggered the development of localization solutions that jointly estimate the robot position and the position of elements in the environment, referred to as Simultaneous Localization and Mapping (SLAM) approaches. To handle the noise inherent of the data coming from the sensors, SLAM solutions are implemented in a probabilistic framework. First developments were based on Extended Kalman Filters, while a more recent developments use probabilistic graphical models to model the estimation problem and solve it through optimization. This thesis exploits the latter approach to develop two distinct techniques for autonomous ground vehicles: oneusing monocular vision, the other one using LIDAR. The lack of depth information in camera images has fostered the use of specific landmark parametrizations that isolate the unknown depth in one variable, concentrating its large uncertainty into a single parameter. One of these parametrizations, named Parallax Angle Parametrization, was originally introduced in the context of the Bundle Adjustment problem, that processes all the gathered data in a single global optimization step. We present how to exploit this parametrization in an incremental graph-based SLAM approach in which robot motion measures are also incorporated. LIDAR sensors can be used to build odometry-like solutions for localization by sequentially registering the point clouds acquired along a robot trajectory. We define a graphical model layer on top of a LIDAR odometry layer, that uses the Iterative Closest Points (ICP) algorithm as registration technique. Reference frames are defined along the robot trajectory, and ICP results are used to build a pose graph, used to solve an optimization problem that enables the correction of the robot trajectory and the environment map upon loop closures. After an introduction to the theory of graphical models applied to SLAM problem, the manuscript depicts these two approaches. Simulated and experimental results illustrate the developments throughout the manuscript, using classic and in-house datasets.

Multi-label classification consists of learning a function that is capable of mapping an object to a set of relevant labels. It has applications such as the association of genes with biological functions, semantic classification of scenes and text categorization. Traditional classification (i.e., single-label) is therefore a particular case of multi-label classification in which each object is associated with exactly one label. A successful approach to constructing classifiers is to obtain a probabilistic model of the relation between object attributes and labels. This model can then be used to classify objects, finding the most likely prediction by computing the marginal probability or the most probable explanation (MPE) of the labels given the attributes. Depending on the probabilistic models family chosen, such inferences may be intractable when the number of labels is large. Sum-Product Networks (SPN) are deep probabilistic models, that allow tractable marginal inference. Nevertheless, as with many other probabilistic models, performing MPE inference is NP- hard. Although, SPNs have already been used successfully for traditional classification tasks (i.e. single-label), there is no in-depth investigation on the use of SPNs for Multi-Label classification. In this work we investigate the use of SPNs for Multi-Label classification. We compare several algorithms for learning SPNs combined with different proposed approaches for classification. We show that SPN-based multi-label classifiers are competitive against state-of-the-art classifiers, such as Random k-Labelsets with Support Vector Machine and MPE inference on CutNets, in a collection of benchmark datasets. / A classificação Multi-Rótulo consiste em aprender uma função que seja capaz de mapear um objeto para um conjunto de rótulos relevantes. Ela possui aplicações como associação de genes com funções biológicas, classificação semântica de cenas e categorização de texto. A classificação tradicional, de rótulo único é, portanto, um caso particular da Classificação Multi-Rótulo, onde cada objeto está associado com exatamente um rótulo. Uma abordagem bem sucedida para classificação é obter um modelo probabilístico da relação entre atributos do objeto e rótulos. Esse modelo pode então ser usado para classificar objetos, encon- trando a predição mais provável por meio da probabilidade marginal ou a explicação mais provavél dos rótulos dados os atributos. Dependendo da família de modelos probabilísticos escolhidos, tais inferências podem ser intratáveis quando o número de rótulos é grande. As redes Soma-Produto (SPN, do inglês Sum Product Network) são modelos probabilísticos profundos, que permitem inferência marginal tratável. No entanto, como em muitos outros modelos probabilísticos, a inferência da explicação mais provavél é NP-difícil. Embora SPNs já tenham sido usadas com sucesso para tarefas de classificação tradicionais, não existe investigação aprofundada no uso de SPNs para classificação Multi-Rótulo. Neste trabalho, investigamos o uso de SPNs para classificação Multi-Rótulo. Comparamos vários algoritmos de aprendizado de SPNs combinados com diferentes abordagens propostos para classi- ficação. Mostramos que os classificadores Multi-Rótulos baseados em SPN são competitivos contra classificadores estado-da-arte, como Random k-Labelsets usando Máquinas de Suporte Vetorial e inferência exata da explicação mais provavél em CutNets, em uma coleção de conjuntos de dados de referência.

Modeling complex cognitive phenomena is a challenging task, especially when it is required to account for the functioning of a cognitive system interacting with an uncertain and changing environment. Psychometrics offers a heterogeneous corpus of computational tools to infer latent cognitive constructs from the observation of behavioral outcomes. However, there is not an explicit consensus regarding the optimal way to properly take into account the intrinsic dynamic properties of the environment, as well as the dynamic nature of cognitive states. In the present dissertation, we explore the potentials of relying on discrete state dynamic models to formally account for the unfolding of cognitive sub-processes in changing task environments. In particular, we propose Probabilistic Graphical Models (PGMs) as an ideal and unifying mathematical language to represent cognitive dynamics as structured graphs codifying (causal) relationships between cognitive sub-components which unfolds in discrete time. We propose several works demonstrating the advantage and the representational power of such a modeling framework, by providing dynamic models of cognition specified according to different levels of abstraction.

<p>One in nine women is expected to be diagnosed with breast cancer during her life. In 2013, an estimated 23, 800 Canadian women will be diagnosed with breast cancer and 5, 000 will die of it. Making decisions about the treatment for a patient is difficult since it depends on various clinical features, genomic factors, and pathological and cellular classification of a tumor.</p> <p>In this research, we propose a probabilistic graphical model for prognosis and diagnosis of breast cancer that can help medical doctors make better decisions about the best treatment for a patient. Probabilistic graphical models are suitable for making decisions under uncertainty from big data with missing attributes and noisy evidence.</p> <p>Using the proposed model, we may enter the results of different tests (e.g. estrogen and progesterone receptor test and HER2/neu test), microarray data, and clinical traits (e.g. woman's age, general health, menopausal status, stage of cancer, and size of the tumor) to the model and answer to following questions. How likely is it that the cancer will extend in the body (distant metastasis)? What is the chance of survival? How likely is that the cancer comes back (local or regional recurrence)? How promising is a treatment? For example, how likely metastasis is and how likely recurrence is for a new patient, if certain treatment e.g. surgical removal, radiation therapy, hormone therapy, or chemotherapy is applied. We can also classify various types of breast cancers using this model.</p> <p>Previous work mostly relied on clinical data. In our opinion, since cancer is a genetic disease, the integration of the genomic (microarray) and clinical data can improve the accuracy of the model for prognosis and diagnosis. However, increasing the number of variables may lead to poor results due to the curse of dimensionality dilemma and small sample size problem. The microarray data is high dimensional. It consists of around 25, 000 variables per patient. Moreover, structure learning and parameter learning for probabilistic graphical models require a significant amount of computations. The number of possible structures is also super-exponential with respect to the number of variables. For instance, there are more than 10^18 possible structures with just 10 variables.</p> <p>We address these problems by applying manifold learning and dimensionality reduction techniques to improve the accuracy of the model. Extensive experiments using real-world data sets such as METRIC and NKI show the accuracy of the proposed method for classification and predicting certain events, like recurrence and metastasis.</p> / Master of Science (MSc)

Estimation of Distribution Algorithms (EDAs) belong to the class of population based optimisation algorithms. They are motivated by the idea of discovering and exploiting the interaction between variables in the solution. They estimate a probability distribution from population of solutions, and sample it to generate the next population. Many EDAs use probabilistic graphical modelling techniques for this purpose. In particular, directed graphical models (Bayesian networks) have been widely used in EDA. This thesis proposes an undirected graphical model (Markov Random Field (MRF)) approach to estimate and sample the distribution in EDAs. The interaction between variables in the solution is modelled as an undirected graph and the joint probability of a solution is factorised as a Gibbs distribution. The thesis describes a model of fitness function that approximates the energy in the Gibbs distribution, and shows how this model can be fitted to a population of solutions to estimate the parameters of the MRF. The estimated MRF is then sampled to generate the next population. This approach is applied to estimation of distribution in a general framework of an EDA, called Distribution Estimation using Markov Random Fields (DEUM). The thesis then proposes several variants of DEUM using different sampling techniques and tests their performance on a range of optimisation problems. The results show that, for most of the tested problems, the DEUM algorithms significantly outperform other EDAs, both in terms of number of fitness evaluations and the quality of the solutions found by them. There are two main explanations for the success of DEUM algorithms. Firstly, DEUM builds a model of fitness function to approximate the MRF. This contrasts with other EDAs, which build a model of selected solutions. This allows DEUM to use fitness in variation part of the evolution. Secondly, DEUM exploits the temperature coefficient in the Gibbs distribution to regulate the behaviour of the algorithm. In particular, with higher temperature, the distribution is closer to being uniform and with lower temperature it concentrates near some global optima. This gives DEUM an explicit control over the convergence of the algorithm, resulting in better optimisation.

Causal inference may help us to understand the underlying mechanisms and the risk factors of diseases. In Genetics, it is crucial to understand how the connectivity among variables is influenced by genetic and environmental factors. Family data have proven to be useful in elucidating genetic and environmental influences, however, few existing approaches are able of addressing structure learning of probabilistic graphical models (PGMs) and family data analysis jointly. We propose methodologies for learning, from observational Gaussian family data, the most likely PGM and its decomposition into genetic and environmental components. They were evaluated by a simulation study and applied to the Genetic Analysis Workshop 13 simulated data, which mimic the real Framingham Heart Study data, and to the metabolic syndrome phenotypes from the Baependi Heart Study. In neuroscience, one challenge consists in identifying interactions between functional brain networks (FBNs) - graphs. We propose a method to identify Granger causality among FBNs. We show the statistical power of the proposed method by simulations and its usefulness by two applications: the identification of Granger causality between the FBNs of two musicians playing a violin duo, and the identification of a differential connectivity from the right to the left brain hemispheres of autistic subjects. / Inferência causal pode nos ajudar a compreender melhor as relações de dependência direta entre variáveis e, assim, a identificar fatores de riscos de doenças. Em Genética, a análise de dados agrupados em famílias permite investigar influências genéticas e ambientais nas relações entre as variáveis. Neste trabalho, nós propomos métodos para aprender, a partir de dados Gaussianos agrupados em famílias, o mais provável modelo gráfico probabilístico (dirigido ou não dirigido) e também sua decomposição em dois componentes: genético e ambiental. Os métodos foram avaliados por simulações e aplicados tanto aos dados simulados do Genetic Analysis Workshop 13, que imitam características dos dados do Framingham Heart Study, como aos dados da síndrome metabólica do estudo Corações de Baependi. Em Neurociência, um desafio consiste em identificar interações entre redes funcionais cerebrais - grafos. Nós propomos um método que identifica causalidade de Granger entre grafos e, por meio de simulações, mostramos que o método tem alto poder estatístico. Além disso, mostramos sua utilidade por meio de duas aplicações: 1) identificação de causalidade de Granger entre as redes cerebrais de dois músicos enquanto tocam um dueto de violino e 2) identificação de conectividade diferencial do hemisfério cerebral direito para o esquerdo em indivíduos autistas.

[pt] O crescimento da população do planeta tem aumentado continuamente a demanda por produtos agrícolas. Assim, a informação quanto a áreas cultivadas e estimativas de produção se tornam cada vez mais importantes. Técnicas baseadas em imagens satelitais constituem uma das opções mais atrativas para o monitoramento agrícola sobre grandes áreas. A maior parte dos trabalhos científicos voltados a esta aplicação foram desenvolvidos para regiões temperadas do planeta, que apresentam um dinâmica muito mais simples da que se tem em regiões tropicais. Neste contexto, a presente tese propõe um novo método automático baseado em Campos Aleatórios Condicionais (CRF) para o reconhecimento de culturas agrícolas em regiões tropicais a partir de sequências de imagens multi-temporais e multiresolução produzidas por diferentes sensores orbitais. Experimentos foram realizados para validar diversas variantes do método proposto. Utilizaramse bases de dados públicas de duas regiões do Brasil que compreendem sequências de imagens óticas e de radar com diferentes resoluções espaciais. Os experimentos realizados demonstraram que o método proposto atingiu acurácias maiores do que métodos baseados em uma única imagem ou sensor. Particularmente, notou-se a redução do efeito sal-e-pimenta nos mapas gerados devido, mormente, à capacidade do método de capturar informação contextual. / [en] The earth population growth has continuously increased the demand for agricultural production. Thus, acreage and crop yield information become increasingly important. Techniques based on satellite images are one of the most attractive options for agricultural monitoring over large areas. Most of the scientific works on this application were developed for temperate regions of the planet, which present a much simpler dynamics than those in tropical regions. In this context, the present thesis proposes a new automatic method based on Conditional Random Fields (CRF) for the crop recognition in tropical regions from multi-temporal and multi-resolution image sequences from orbital multi-sensors. Experiments were performed to validate several variants of the proposed method. We used public databases from two regions of Brazil that comprise sequences of optical and radar images with different spatial resolutions. The experiments demonstrated that the proposed method achieved a higher accuracy than methods based on a single image or sensor. Particularly, the reduction of the salt-and-pepper effect in the generated maps was noticed due, mainly, to the capacity of the method to capture contextual information.

Le texte est l'une des sources d'informations les plus répandues et les plus persistantes. L'analyse de contenu du texte se réfère à des méthodes d'étude et de récupération d'informations à partir de documents. Aujourd'hui, avec une quantité de texte disponible en ligne toujours croissante l'analyse de contenu du texte revêt une grande importance parce qu' elle permet une variété d'applications. À cette fin, les méthodes d'apprentissage de la représentation sans supervision telles que les modèles thématiques et les word embeddings constituent des outils importants.L'objectif de cette dissertation est d'étudier et de relever des défis dans ce domaine.Dans la première partie de la thèse, nous nous concentrons sur les modèles thématiques et plus précisément sur la manière d'incorporer des informations antérieures sur la structure du texte à ces modèles.Les modèles de sujets sont basés sur le principe du sac-de-mots et, par conséquent, les mots sont échangeables. Bien que cette hypothèse profite les calculs des probabilités conditionnelles, cela entraîne une perte d'information.Pour éviter cette limitation, nous proposons deux mécanismes qui étendent les modèles de sujets en intégrant leur connaissance de la structure du texte. Nous supposons que les documents sont répartis dans des segments de texte cohérents. Le premier mécanisme attribue le même sujet aux mots d'un segment. La seconde, capitalise sur les propriétés de copulas, un outil principalement utilisé dans les domaines de l'économie et de la gestion des risques, qui sert à modéliser les distributions communes de densité de probabilité des variables aléatoires tout en n'accédant qu'à leurs marginaux.La deuxième partie de la thèse explore les modèles de sujets bilingues pour les collections comparables avec des alignements de documents explicites. En règle générale, une collection de documents pour ces modèles se présente sous la forme de paires de documents comparables. Les documents d'une paire sont écrits dans différentes langues et sont thématiquement similaires. À moins de traductions, les documents d'une paire sont semblables dans une certaine mesure seulement. Pendant ce temps, les modèles de sujets représentatifs supposent que les documents ont des distributions thématiques identiques, ce qui constitue une hypothèse forte et limitante. Pour le surmonter, nous proposons de nouveaux modèles thématiques bilingues qui intègrent la notion de similitude interlingue des documents qui constituent les paires dans leurs processus générateurs et d'inférence.La dernière partie de la thèse porte sur l'utilisation d'embeddings de mots et de réseaux de neurones pour trois applications d'exploration de texte. Tout d'abord, nous abordons la classification du document polylinguistique où nous soutenons que les traductions d'un document peuvent être utilisées pour enrichir sa représentation. À l'aide d'un codeur automatique pour obtenir ces représentations de documents robustes, nous démontrons des améliorations dans la tâche de classification de documents multi-classes. Deuxièmement, nous explorons la classification des tweets à plusieurs tâches en soutenant que, en formant conjointement des systèmes de classification utilisant des tâches corrélées, on peut améliorer la performance obtenue. À cette fin, nous montrons comment réaliser des performances de pointe sur une tâche de classification du sentiment en utilisant des réseaux neuronaux récurrents. La troisième application que nous explorons est la récupération d'informations entre langues. Compte tenu d'un document écrit dans une langue, la tâche consiste à récupérer les documents les plus similaires à partir d'un ensemble de documents écrits dans une autre langue. Dans cette ligne de recherche, nous montrons qu'en adaptant le problème du transport pour la tâche d'estimation des distances documentaires, on peut obtenir des améliorations importantes. / Text is one of the most pervasive and persistent sources of information. Content analysis of text in its broad sense refers to methods for studying and retrieving information from documents. Nowadays, with the ever increasing amounts of text becoming available online is several languages and different styles, content analysis of text is of tremendous importance as it enables a variety of applications. To this end, unsupervised representation learning methods such as topic models and word embeddings constitute prominent tools.The goal of this dissertation is to study and address challengingproblems in this area, focusing on both the design of novel text miningalgorithms and tools, as well as on studying how these tools can be applied to text collections written in a single or several languages.In the first part of the thesis we focus on topic models and more precisely on how to incorporate prior information of text structure to such models.Topic models are built on the premise of bag-of-words, and therefore words are exchangeable. While this assumption benefits the calculations of the conditional probabilities it results in loss of information.To overcome this limitation we propose two mechanisms that extend topic models by integrating knowledge of text structure to them. We assume that the documents are partitioned in thematically coherent text segments. The first mechanism assigns the same topic to the words of a segment. The second, capitalizes on the properties of copulas, a tool mainly used in the fields of economics and risk management that is used to model the joint probability density distributions of random variables while having access only to their marginals.The second part of the thesis explores bilingual topic models for comparable corpora with explicit document alignments. Typically, a document collection for such models is in the form of comparable document pairs. The documents of a pair are written in different languages and are thematically similar. Unless translations, the documents of a pair are similar to some extent only. Meanwhile, representative topic models assume that the documents have identical topic distributions, which is a strong and limiting assumption. To overcome it we propose novel bilingual topic models that incorporate the notion of cross-lingual similarity of the documents that constitute the pairs in their generative and inference processes. Calculating this cross-lingual document similarity is a task on itself, which we propose to address using cross-lingual word embeddings.The last part of the thesis concerns the use of word embeddings and neural networks for three text mining applications. First, we discuss polylingual document classification where we argue that translations of a document can be used to enrich its representation. Using an auto-encoder to obtain these robust document representations we demonstrate improvements in the task of multi-class document classification. Second, we explore multi-task sentiment classification of tweets arguing that by jointly training classification systems using correlated tasks can improve the obtained performance. To this end we show how can achieve state-of-the-art performance on a sentiment classification task using recurrent neural networks. The third application we explore is cross-lingual information retrieval. Given a document written in one language, the task consists in retrieving the most similar documents from a pool of documents written in another language. In this line of research, we show that by adapting the transportation problem for the task of estimating document distances one can achieve important improvements.