De nos jours, il est très fréquent de représenter un système en termes de relations entre objets. Parmi les applications les plus courantes de telles données relationnelles, se situent les systèmes de recommandation (RS), qui traitent généralement des relations entre utilisateurs et items à recommander. Les modèles relationnels probabilistes (PRM) sont un bon choix pour la modélisation des dépendances probabilistes entre ces objets. Une tendance croissante dans les systèmes de recommandation est de rajouter une dimension spatiale à ces objets, que ce soient les utilisateurs, ou les items. Cette thèse porte sur l’intersection peu explorée de trois domaines connexes - modèles probabilistes relationnels (et comment apprendre les dépendances probabilistes entre attributs d’une base de données relationnelles), les données spatiales et les systèmes de recommandation. La première contribution de cette thèse porte sur le chevauchement des PRM et des systèmes de recommandation. Nous avons proposé un modèle de recommandation à base de PRM capable de faire des recommandations à partir des requêtes des utilisateurs, mais sans profils d’utilisateurs, traitant ainsi le problème du démarrage à froid. Notre deuxième contribution aborde le problème de l’intégration de l’information spatiale dans un PRM. / Nowadays it is very common to represent a system in terms of relationships between objects. One of the common applications of such relational data is Recommender System (RS), which usually deals with the relationships between users and items. Probabilistic Relational Models (PRMs) can be a good choice for modeling probabilistic dependencies between such objects. A growing trend in recommender systems is to add spatial dimensions to these objects, and make recommendations considering the location of users and/or items. This thesis deals with the (not much explored) intersection of three related fields – Probabilistic Relational Models (a method to learn probabilistic models from relational data), spatial data (often used in relational settings), and recommender systems (which deal with relational data). The first contribution of this thesis deals with the overlapping of PRM and recommender systems. We have proposed a PRM-based personalized recommender system that is capable of making recommendations from user queries in cold-start systems without user profiles. Our second contribution addresses the problem of integrating spatial information into a PRM.

Les fibres optiques légèrement multimodes (FMF) sont une classe de fibres optiques multimodes. Dans le domaine de la télécommunication, chaque mode d’une FMF peut être utilisé comme un canal indépendant de transmission, ainsi, des débits bien supérieurs aux fibres optiques conventionnelles pourrait être atteint. Cependant, l'un des problèmes à l'utilisation de ce type de fibre est le couplage de mode et la dispersion de mode qui dégradent les performances de transmission. Cette thèse entre dans le cadre de l’étude du couplage de mode de différents profils de fibre. Ce travail se décompose en deux axes : i) la modélisation du couplage de mode qui est effectué en considérant la fibre comme une concaténation de plusieurs segments courbés, dont chaque segment est associé à un rayon de courbure aléatoire. ii) La mesure du couplage localisée qui est basé sur la méthode expérimentale A-S2. A l’aide de ce modèle et de ces mesures, nous pouvons confirmer l’absence ou la présence de couplage d’une fibre exposée à des micro-courbures ou à une perturbation aléatoire. / Few multimode fiber (FMF) is a class of multimode fiber. Each mode of a FMF is considered as an independent transmission channel in telecommunication field. Thus, much higher rates than conventional optical fibers could be achieved. However, this fiber can present the mode coupling and mode dispersion that can degrade the transmission performance. The mode coupling over different fiber profiles is studied in this thesis. This work is divided into two parts: i) the modeling of the mode coupling considering the fiber as a concatenation of several curved segments, and each piece is associated with a random bending radius R. ii) The measurement of localized coupling which is based on the experimental method A-S2. Using this model and these measurements, we can demonstrate the absence or the presence of coupling of a fiber exposed to micro-curvatures or a random perturbation.

The role of spatial reasoning in the development of systems in the domain of Artificial Intelligence is increasing. One particular approach, qualitative spatial reasoning, investigates the usage of abstract representation to facilitate the representation of and the reasoning with spatial information.This thesis investigates the usage of intervals along global axes as the under-lying representational and reasoning mechanism for a spatial reasoning system. Aspects that are unique to representing spatial information (flow and multi-dimensionality) are used to provide a method for classifying relations between objects at multiple levels of granularity. The combination of these two mechanisms (intervals and classification) provide the basis for the development of a querying system that allows qualitative queries about object relations in multi-dimensional space to be performed upon the representation.The second issue examined by this thesis is the problem of representing intervals when all the interval relations may not be known precisely. A three part solution is proposed. The first shows how the simplest situation, where all relations are explicit and primitive, can be represented and integrated with the above mentioned querying system. The second situation demonstrates how, for interval relations that are primitive but are not all explicitly known, an effective point based representation may be constructed. Finally, when relations between intervals are disjunctions of possible primitive interval relations, a representation is presented which allows solutions to queries to be constructed from consistent data.Our contribution is two-fold:1. a method of classifying the spatial relations and the means of querying these relations;2. a process of efficiently representing incomplete interval information and the means of efficiently querying this information.The work presented ++ / in this thesis demonstrates the utility of a multi-dimensional qualitative spatial reasoning system based upon intervals. It also demonstrates how an interval representation may be constructed for datasets that have variable levels of information about relationships between intervals represented in the dataset.

Nowadays there are a large number of applications, such as digital library information retrieval, business data analysis, CAD/CAM, multimedia applications with images and sound, real-time process control and scientific computation, with data sets about gigabytes, terabytes or even petabytes. Because data distributions are too large to be stored accurately, maintaining compact and accurate summarized information about underlying data is of crucial important. The summarizing problem for Level 1 (disjoint and non-disjoint) topological relationship has been well studied for the past few years. However the spatial database users are often interested in a much richer set of spatial relations such as contains. Little work has been done on summarization for Level 2 topological relationship which includes contains, contained, overlap, equal and disjoint relations. We study the problem of effective summatization to represent the underlying data distribution to answer window queries for Level 2 topological relationship. Cell-density based approach has been demonstrated as an effective way to this problem. But the challenges are the accuracy of the results and the storage space required which should be linearly proportional to the number of cells to be practical. In this thesis, we present several novel techniques to effectively construct cell density based spatial histograms. Based on the framework proposed, exact results could be obtained in constant time for aligned window queries. To minimize the storage space of the framework, an approximate algorithm with the approximate ratio 19/12 is presented, while the problem is shown NP-hard generally. Because the framework requires only a storage space linearly proportional to the number of cells, it is practical for many popular real datasets. To conform to a limited storage space, effective histogram construction and query algorithms are proposed which can provide approximate results but with high accuracy. The problem for non-aligned window queries is also investigated and techniques of un-even partitioned space are developed to support non-aligned window queries. Finally, we extend our techniques to 3D space. Our extensive experiments against both synthetic and real world datasets demonstrate the efficiency of the algorithms developed in this thesis.

There are many existing methods for capturing three dimensional data from two dimensional images. Methods based on images captured from multiple view-points require solving the correspondence problem: establishing which points in each image represent the same points in the scene. Most attempts at solving the correspondence problem require carefully controlled lighting and reference points within the scene. A new method captures many consecutive images to form a dense spatiotemporal volume as the camera-or scene-undergoes controlled motion. Feature points in the scene move along predictable paths within this volume. Analysing the exact motion of features determines their three dimensional position in the scene.

<p>Denna studie har som syfte att undersöka hur den spatiala organisationen av artefakter påverkar överrapportering mellan sjukvårdspersonal på en intensivvårdsavdelning. Syftet äratt kunna förstå överrapporteringsprocessen mellan sjuksköterskor och de artefakter de har tillsitt förfogande och att ge förslag på strukturförbättringar så att dessa överrapporteringar kan effektiviseras både tid- och energimässigt. Det leder till studiens andra syfte, nämligen att ge förslag på hur en gemensam arbetsyta vid överrapporteringar, ett så kallat worktable, skulle kunna arbetas fram. Resultatet av denna studie visar att det inte finns en given struktur för hur överrapporteringar ska genomföras och därför finns det inte heller en bestämd ordning över artefakters spatiala organisation. I slutet av denna rapport ges det förslag på vilka funktioner ett framtida gemensamt arbetsbord borde innefatta.</p>

Thesis (Ph. D.)--University of Chicago, Dept. of Statistics, June 1999. / Includes bibliographical references. Also available on the Internet.

The purpose of this study was to describe the herd antibody status to Salmonella Typhimurium as estimated from co-mingled milk samples and to describe the resulting geographical patterns found in Texas dairy herds. Bulk tank milk samples were collected from 852 Grade A dairies throughout Texas during the summer of 2001. An indirect enzyme-linked immunosorbent assay (ELISA) using S. Typhimurium lipopolysaccharide was performed with signal to noise ratios calculated for each sample. The ELISA ratio was used in fitting a theoretical variogram and kriging was used to develop a predicted surface for these ratios in Texas. A spatial process with areas of higher risk located in the panhandle and near Waller County was apparent. Lower risk areas included Atascosa, Cooke, Collin, Titus, Comanche and Cherokee Counties. Subsets representing large dairy sheds in northeast Texas, the Erath County area, and the Hopkins County area were also evaluated individually. Each result illustrated a spatial process with areas of low and high ELISA ratio predictions. Cluster analysis was performed for the entire state with cases defined as herds having milk ELISA ratios greater than or equal to 1.8. Using this cutoff, the prevalence of herds with positive bulk tank milk ELISAs was 4.3%. Significant clustering of cases was demonstrated by the Cuzick and Edward's test. The spatial scan statistic then identified the two most likely clusters located in and near the Texas Panhandle. This study demonstrated that the distribution of S. Typhimurium antibodies in bulk tank milk in Texas is describable by a spatial process. Knowledge of this process will help elucidate geospatial influences on the presence of S. Typhimurium in dairy herds and enhance our understanding of the epidemiology of salmonellosis.

The objective of the present work is to propose a novel method to extract a stable feature set representative of image content. Each image is represented by a linear combination of fractal orthonormal basis vectors. The mapping coefficients of an image projected onto each orthonormal basis constitute the feature vector. The set of orthonormal basis vectors are generated by utilizing fractal iterative function through target and domain blocks mapping. The distance measure remains consistent, i.e., isometric embedded, between any image pairs before and after the projection onto orthonormal axes. Not only similar images generate points close to each other in the feature space, but also dissimilar ones produce feature points far apart. The above statements are logically equivalent to that distant feature points are guaranteed to map to images with dissimilar contents, while close feature points correspond to similar images. Therefore, utilizing coefficients derived from the proposed linear combination of fractal orthonormal basis as key to search image database will retrieve similar images, while at the same time exclude dissimilar ones. The coefficients associated with each image can be later used to reconstruct the original. The content-based query is performed in the compressed domain. This approach is efficient for content-based query. Scaling, rotational, translation, mirroring and horizontal/vertical flipping variations of a query image are also supported. A symbolic image database system is a system in which a large amount of image data and their related information are represented by both symbolic images and physical images. How to perceive spatial relationships among the components in a symbolic image is an important criterion to find a match between the symbolic image of the scene object and the one being store as a modal in the symbolic image database. Spatial reasoning techniques have been applied to pictorial database, in particular those using 2D strings as an index representation have been successful. In most of the previous approaches for iconic indexing, for simplifying the concerns, they apply the MBR (Minimum bounding rectangle) of two objects to define the spatial relationship between them. Multiple instance learning algorithms provide ways for computer program to improve automatically with experience. Most images are inherently ambiguous disseminators of information. Unfortunately, interfaces to image databases normally involve the user giving the system ambiguous queries. By treating each query as a Multiple-Instance example, we make the ambiguity in each image explicit. In addition, by receiving several positive and negative examples, the system can learn what the user desires. Using the learned concept, the system returns images from the database that are close to that concept. In this project, we propose to apply the Multiple-Instance learning model by deriving the projection vector of fractal orthonormal bases for a small number of training images to learn what images from database are of interest to the user.

none