Many applications use multiple cameras to simultaneously capture imagery of a scene from different vantage points on a rigid, moving camera system over time. Multiple cameras often provide unique viewing angles but also additional levels of detail of a scene at different spatio-temporal resolutions. However, in order to benefit from this added information the sources must be temporally aligned. As a result of cost and physical limitations it is often impractical to synchronize these sources via an external clock device. Most methods attempt synchronization through the recovery of a constant scale factor and offset with respect to time. This limits the generality of such alignment solutions. We present an unsupervised method that utilizes a content-based clustering mechanism in order to temporally align multiple non-synchronized image sequences of different and varying spatio-temporal resolutions. We show that the use of temporal constraints and dynamic programming adds robustness to changes in capture rates, field of view, and resolution.

Technical disciplines are evolving rapidly leading to changes in their associated vocabularies. Confusion in interdisciplinary communication occurs due to this evolving terminology. Two causes of confusion are multiple definitions (overloaded terms) and synonymous terms. The formal names for these two problems are polysemy and synonymy. Termediator-I, a web application built on top of a collection of glossaries, uses definition count as a measure of term confusion. This tool was an attempt to identify confusing cross-disciplinary terms. As more glossaries were added to the collection, this measure became ineffective. This thesis provides a measure of term polysemy. Term polysemy is effectively measured by semantically clustering the text concepts, or definitions, of each term and counting the number of resulting clusters. Hierarchical clustering uses a measure of proximity between the text concepts. Three such measures are evaluated: cosine similarity, latent semantic indexing, and latent Dirichlet allocation. Two linkage types, for determining cluster proximity during the hierarchical clustering process, are also evaluated: complete linkage and average linkage. Crowdsourcing through a web application was unsuccessfully attempted to obtain a viable clustering threshold by public consensus. An alternate metric of polysemy, convergence value, is identified and tested as a viable clustering threshold. Six resulting lists of terms ranked by cluster count based on convergence values are generated, one for each similarity measure and linkage type combination. Each combination produces a competitive list, and no clear combination can be determined as superior. Semantic clustering successfully identifies polysemous terms, but each similarity measure and linkage type combination provides slightly different results.

This thesis provides a new hierarchical clustering algorithm for graphs, named Paris, which can be interpreted through the modularity score and its resolution parameter. The algorithm is agglomerative and based on a simple distance between clusters induced by the probability of sampling node pairs. It tries to approximate the optimal partitions with respect to the modularity score at any resolution in one run. In addition to the Paris hierarchical algorithm, this thesis proposes four algorithms that compute rankings of the sharpest clusters, clusterings and resolutions by processing the hierarchy output by Paris. These algorithms are based on a new measure of stability for clusterings, named sharp-score. Key outcomes of these four algorithms are the possibility to rank clusters, detect sharpest clusterings scale, go beyond the resolution limit and detect relevant resolutions. All these algorithms have been tested on both synthetic and real datasets to illustrate the efficiency of their approaches. / Denna avhandling ger en ny hierarkisk klusteralgoritm för grafer, som heter Paris, vilket kan tolkas av modularitetsresultatet och dess upplösningsparameter. Algoritmen är agglomerativ och är baserad på ett enda avstånd mellan kluster som induceras av sannolikheten för sampling av nodpar. Det försöker att approximera de optimala partitionerna vid vilken upplösning som helst i en körning. Förutom en hierarkisk algoritm föreslår denna avhandling fyra algoritmer som beräknar rankningar av de bästa grupperna, kluster och resolutioner genom att bearbeta hierarkiproduktionen i Paris. Dessa algoritmer bygger på ett nytt koncept av klusterstabilitet, kallad sharpscore. Viktiga resultat av dessa fyra algoritmer är förmågan att rangordna kluster, upptäcka bästa klusterskala, gå utöver upplösningsgränsen och upptäcka de mest relevanta resolutionerna. Alla dessa algoritmer har testats på både syntetiska och verkliga datamängder för att illustrera effektiviteten i deras metoder.

Vehicle Routing Problems are optimization problems centered around determining optimal travel routes for a fleet of vehicles to visit a set of nodes. Optimality is evaluated with regard to some desired quality of the solution, such as time-minimizing or cost-minimizing. There are many established solution methods which makes it meaningful to compare their performance. This thesis aims to investigate how the performances of various solution methods is affected by varying certain problem parameters. Problem characteristics such as the number of customers, vehicle capacity, and customer demand are investigated. The aim was approached by dividing the problem into two subproblems: distributing the nodes into suitable clusters, and finding the shortest route within each cluster. Results were produced by solving simulated sets of customers for different parameter values with different clustering methods, namely sweep, k-means and hierarchical clustering. Although the model required simplifications to facilitate the implementation, theresults provided some significant findings. The thesis concludes that for large vehicle capacity in relation to demand, sweep clustering is the preferred method. Whereas for smaller vehicles, the other two methods perform better.

This thesis presents a method for integrating heterogeneous gene/protein datasets at the functional level based on Gene Ontology term similarity. Often biologists want to integrate heterogeneous data sets obtain from different biological samples. A major challenge in this process is how to link the heterogeneous datasets. Currently, the most common approach is to link them through common reference database identifiers which tend to result in small number of matching identifiers. This is due to lack of standard accession schemes. Due to this problem, biologists may not recognize the underlying biological phenomena revealed by a combination of the data but by each data set individually. We discuss an approach for integrating heterogeneous datasets by computing the similarity among them based on the similarity of their GO annotations. Then we group the genes and/or proteins with similar annotations by applying a hierarchical clustering algorithm. The results demonstrate a more comprehensive understanding of the biological processes involved.

No description available.

No description available.

With the rapid development of modern high-throughput technologies, scientists can now collect high-dimensional complex data in different forms, such as medical images, genomics measurements. However, acquisition of more data does not automatically lead to better knowledge discovery. One needs efficient and reliable analytical tools to extract useful information from complex datasets. The main objective of this dissertation is to develop innovative Bayesian methodologies to enable effective and efficient knowledge discovery from complex high-dimensional data. It contains two parts—the development of computationally efficient functional mixed models and the modeling of data heterogeneity via Dirichlet Diffusion Tree. The first part focuses on tackling the computational bottleneck in Bayesian functional mixed models. We propose a computational framework called variational functional mixed model (VFMM). This new method facilitates efficient data compression and high-performance computing in basis space. We also propose a new multiple testing procedure in basis space, which can be used to detect significant local regions. The effectiveness of the proposed model is demonstrated through two datasets, a mass spectrometry dataset in a cancer study and a neuroimaging dataset in an Alzheimer's disease study. The second part is about modeling data heterogeneity by using Dirichlet Diffusion Trees. We propose a Bayesian latent tree model that incorporates covariates of subjects to characterize the heterogeneity and uncover the latent tree structure underlying data. This innovative model may reveal the hierarchical evolution process through branch structures and estimate systematic differences between groups of samples. We demonstrate the effectiveness of the model through the simulation study and a brain tumor real data. / Doctor of Philosophy / With the rapid development of modern high-throughput technologies, scientists can now collect high-dimensional data in different forms, such as engineering signals, medical images, and genomics measurements. However, acquisition of such data does not automatically lead to efficient knowledge discovery. The main objective of this dissertation is to develop novel Bayesian methods to extract useful knowledge from complex high-dimensional data. It has two parts—the development of an ultra-fast functional mixed model and the modeling of data heterogeneity via Dirichlet Diffusion Trees. The first part focuses on developing approximate Bayesian methods in functional mixed models to estimate parameters and detect significant regions. Two datasets demonstrate the effectiveness of proposed method—a mass spectrometry dataset in a cancer study and a neuroimaging dataset in an Alzheimer's disease study. The second part focuses on modeling data heterogeneity via Dirichlet Diffusion Trees. The method helps uncover the underlying hierarchical tree structures and estimate systematic differences between the group of samples. We demonstrate the effectiveness of the method through the brain tumor imaging data.

No description available.

Les modèles génétiques de réservoirs sont construits par la simulation des principaux processus de sédimentation dans le temps. En particulier, les modèles tridimensionnels de systèmes chenalisés méandriformes peuvent être construits à partir de trois processus principaux : la migration du chenal, l’aggradation du système et les avulsions, comme c’est réalisé dans le logiciel Flumy pour les environnements fluviatiles. Pour une utilisation opérationnelle, par exemple la simulation d'écoulements, ces simulations doivent être conditionnées aux données d'exploration disponibles (diagraphie de puits, sismique, …). Le travail présenté ici, basé largement sur des développements antérieurs, se concentre sur le conditionnement du modèle Flumy aux données de puits.Deux questions principales ont été examinées au cours de cette thèse. La première concerne la reproduction des données connues aux emplacements des puits. Cela se fait actuellement par une procédure de "conditionnement dynamique" qui consiste à adapter les processus du modèle pendant le déroulement de la simulation. Par exemple, le dépôt de sable aux emplacements de puits est favorisé, lorsque cela est souhaité, par une adaptation des processus de migration ou d'avulsion. Cependant, la manière dont les processus sont adaptés peut générer des effets indésirables et réduire le réalisme du modèle. Une étude approfondie a été réalisée afin d'identifier et d'analyser les impacts indésirables du conditionnement dynamique. Les impacts ont été observés à la fois à l'emplacement des puits et dans tout le modèle de blocs. Des développements ont été réalisés pour améliorer les algorithmes existants.La deuxième question concerne la détermination des paramètres d’entrée du modèle, qui doivent être cohérents avec les données de puits. Un outil spécial est intégré à Flumy - le "Non-Expert User Calculator" (Nexus) - qui permet de définir les paramètres de simulation à partir de trois paramètres clés : la proportion de sable, la profondeur maximale du chenal et l’extension latérale des corps sableux. Cependant, les réservoirs naturels comprennent souvent plusieurs unités stratigraphiques ayant leurs propres caractéristiques géologiques. L'identification de telles unités dans le domaine étudié est d'une importance primordiale avant de lancer une simulation conditionnelle avec des paramètres cohérents pour chaque unité. Une nouvelle méthode de détermination des unités stratigraphiques optimales à partir des données de puits est proposée. Elle est basée sur la Classification géostatistique hiérarchique appliquée à la courbe de proportion verticale (VPC) globale des puits. Les unités stratigraphiques ont pu être détectées à partir d'exemples de données synthétiques et de données de terrain, même lorsque la VPC globale des puits n'était pas visuellement représentative. / Process-based reservoir models are generated by the simulation of the main sedimentation processes in time. In particular, three-dimensional models of meandering channelized systems can be constructed from three main processes: migration of the channel, aggradation of the system and avulsions, as it is performed in Flumy software for fluvial environments. For an operational use, for instance flow simulation, these simulations need to be conditioned to available exploration data (well logging, seismic, …). The work presented here, largely based on previous developments, focuses on the conditioning of the Flumy model to well data.Two main questions have been considered during this thesis. The major one concerns the reproduction of known data at well locations. This is currently done by a "dynamic conditioning" procedure which consists in adapting the model processes while the simulation is running. For instance, the deposition of sand at well locations is favored, when desired, by an adaptation of migration or avulsion processes. However, the way the processes are adapted may generate undesirable effects and could reduce the model realism. A thorough study has been conducted in order to identify and analyze undesirable impacts of the dynamic conditioning. Such impacts were observed to be present both at the location of wells and throughout the block model. Developments have been made in order to improve the existing algorithms.The second question is related to the determination of the input model parameters, which should be consistent with the well data. A special tool is integrated in Flumy – the Non Expert User calculator (Nexus) – which permits to define the simulation parameters set from three key parameters: the sand proportion, the channel maximum depth and the sandbodies lateral extension. However, natural reservoirs often consist in several stratigraphic units with their own geological characteristics. The identification of such units within the studied domain is of prime importance before running a conditional simulation, with consistent parameters for each unit. A new method for determining optimal stratigraphic units from well data is proposed. It is based on the Hierarchical Geostatistical Clustering applied to the well global Vertical Proportion Curve (VPC). Stratigraphic units could be detected from synthetic and field data cases, even when the global well VPC was not visually representative.