The interquartile range (IQR) is used to describe the spread of a distribution. In an introductory statistics course, the IQR might be introduced as simply the “range within which the middle half of the data points lie.” In other words, it is the distance between the two quartiles, IQR = Q3 - Q1. We will compute the population IQR, the expected value, and the variance of the sample IQR for various continuous distributions. In addition, a bootstrap confidence interval for the population IQR will be evaluated.

Discrete Events Simulation (DES) is a powerful tool of modeling and analysis used in different disciplines. DES models require data in order to determine the different parameters that drive the simulations. The literature about DES input data management indicates that the preparation of necessary input data is often a highly manual process, which causes inefficiencies, significant time consumption and a negative user experience. The focus of this research investigation is addressing the manual data collection and processing (MDCAP) problem prevalent in DES projects. This research investigation presents an integrated framework to solve the MDCAP problem by classifying the data needed for DES projects into three generic classes. Such classification permits automating and streamlining the preparation of the data, allowing DES modelers to collect, update, visualize, fit, validate, tally and test data in real-time, by performing intuitive actions. In addition to the proposed theoretical framework, this project introduces an innovative user interface that was programmed based on the ideas of the proposed framework. The interface is called DESI, which stands for Discrete Event Simulation Inputs. The proposed integrated framework to automate DES input data preparation was evaluated against benchmark measures presented in the literature in order to show its positive impact in DES input data management. This research investigation demonstrates that the proposed framework, instantiated by the DESI interface, addresses current gaps in the field, reduces the time devoted to input data management within DES projects and advances the state-of-the-art in DES input data management automation.

The following report chronicles the University of North Texas Wind Turbine Project at Apogee Stadium. The timeline of events will include the feasibility study conducted by and for the university, grant awards from the Texas State Energy Conservation Office to fund the project, and a three-year sample of real time performance data since installation. The purpose of this case study is to compare the energy generation estimates by various stakeholders to the measured energy generation using a new but uniform performance relationship. In order to optimize energy generation in wind turbine generator systems, the most common wind speeds measured at the site should also be the most efficient wind speeds at which the wind turbine can convert the kinetic energy in the wind into mechanical energy and ultimately electrical energy. The tool used to convey this relationship will be a figure plotting the wind speed profile against the efficiency curve of the wind turbine. Applying this relationship tool to the UNT Apogee Stadium wind turbines provided valuable results. The most common wind speeds at Apogee Stadium are not the most efficient wind speed for the turbine. Also, the most common wind speeds were near the lower limit of the wind turbine’s performance parameters. This scenario was evident in both the energy generation predictions as well as the real-time recorded data. This case study will also present the economic analysis of the Apogee Stadium wind turbines using another tool that was not previously used in the feasibility study. The case study concludes with future steps to improve wind turbine performance, and to budget future cost using past, present and future energy savings.

No description available.

Les objectifs de cette thèse visent le développement d’un traitement inédit dans un repère spatiale tridimensionnel, pour le problème de la dynamique de collisions diffusives d’objets macromoléculaires en solution en écoulement hydrodynamique à l'intérieur des pores de largeur variable, soumis aux forces hydrodynamiques, du mouvement brownien et des collisions diffusifs aux parois des pores, en utilisant la modélisation théorique et les simulations numériques. L’approche par simulation numérique est nécessaire car il est extrêmement complexe d’utiliser des outils analytiques à présent pour traiter le problème de ces collisions diffusives aux parois solides. Les algorithmes que nous avons développés et les simulations correspondantes sont suffisamment généraux et avancés pour être directement appliquée à l'étude de la dynamique d'une grande variété de polymère et des particules biologiques dans des solutions diluées sous diverses conditions physiques et hydrodynamiques à l'intérieur des pores. Par ailleurs, les mécanismes conduisant à l'adhésion de nano particules et de particules macromoléculaires sous conditions de non-équilibre, en raison de l'influence contradictoire des collisions mécaniques diffusifs et les forces attractives de Hamaker aux parois solides, sont d'un intérêt majeur. Nous avons donc développé un modèle théorique pour calculer le coefficient de restitution. L'objectif est de quantifier le bilan énergétique pendant le processus de collision diffusive de ces particules aux parois, sous l'influence des forces de répulsion d'une part et les forces attractives de Hamaker. Cela se fait par l'élaboration d'un modèle, basé sur le JKR et les théories d’Hertz, pour tenir compte des pertes d'énergie lors des collisions et des gains d'énergie en raison des interactions Hamaker. L’adhésion arrive si le bilan énergétique le permet. Notre modèle théorique est développé en proposant une approche particulière basée sur le potentiel Hamaker. Nous démontrons ce bilan par le biais d'une équation caractéristique non linéaire pour le coefficient de restitution, et analysons ses propriétés qui déterminent l'adhésion ou non pour diverses conditions physiques initiales. / The purpose of this thesis is to develop a comprehensive model analysis in a three-dimensional spatial frame for the dynamics of molecular particles in dilute colloidal suspensions in solutions flowing inside pores of variable width, subject to hydrodynamic forces, Brownian motion and diffusive collisions at the rough pore boundaries, by using numerical simulations. The approach by simulations is necessary because it is extremely complex to use analytical tools at present to deal with the problem of diffusive collisions of the particles at the solid pore boundaries. The algorithms which we have developed and the corresponding simulations are sufficiently general and refined to be directly applied to the study of the dynamics of a wide variety of polymer and biological particles in dilute solutions under diverse physical and applicable hydrodynamic conditions inside pores. Moreover, the mechanisms leading to the adhesion of particles of nano sizes under what would be non-equilibrium conditions, due to the conflicting influence of the mechanical diffusive collisions and the attractive Hamaker forces at the boundaries, are of major interest. We have hence investigated a theoretical model to calculate the restitution coefficient from basic physical principles. The objective is to quantify the energy balance during the process of a diffusive collision of a nano particle under the influence of the repulsive forces on one hand, and the attractive Hamaker forces acting on the nano particle on the other. This is done by developing a model, based on the JKR and Hertz theories, to account for the energy losses during collisions, and for the energy gains due to the Hamaker interactions. Adhesion becomes an outcome if the energy balance permits this. Our theoretical model is developed by proposing a special analytic approach based on the Hamaker potential. We derive from the theoretical analysis a characteristic nonlinear equation for the restitution coefficient, and analyze its properties which determine under given physical conditions the outcome for adhesion or not.

Disposer de données dans le but de valider ou tester une approche ou un concept est d'une importance primordiale dans beaucoup de domaines différents. Malheureusement, ces données ne sont pas toujours disponibles, sont coûteuses à obtenir, ou bien ne répondent pas à certaines exigences de qualité ce qui les rend inutiles dans certains cas de figure.Un générateur automatique de données est un bon moyen pour obtenir facilement et rapidement des données valides, de différentes tailles, pertinentes et diversifiées. Dans cette thèse, nous proposons une nouvelle approche complète, dirigée par les modèles et basée sur la programmation par contraintes pour la génération de données. / Owning data is useful in many different fields. Data can be used to test and to validate approaches, algorithms and concepts. Unfortunately, data is rarely available, is cost to obtain, or is not adapted to most of cases due to a lack of quality.An automated data generator is a good way to generate quickly and easily data that are valid, in different sizes, likelihood and diverse.In this thesis, we propose a novel and complete model driven approach, based on constraint programming for automated data generation.

Les objectifs de cette thèse visent le développement d'un traitement inédit dans un repère spatiale tridimensionnel, pour le problème de la dynamique de collisions diffusives d'objets macromoléculaires en solution en écoulement hydrodynamique à l'intérieur des pores de largeur variable, soumis aux forces hydrodynamiques, du mouvement brownien et des collisions diffusifs aux parois des pores, en utilisant la modélisation théorique et les simulations numériques. L'approche par simulation numérique est nécessaire car il est extrêmement complexe d'utiliser des outils analytiques à présent pour traiter le problème de ces collisions diffusives aux parois solides. Les algorithmes que nous avons développés et les simulations correspondantes sont suffisamment généraux et avancés pour être directement appliquée à l'étude de la dynamique d'une grande variété de polymère et des particules biologiques dans des solutions diluées sous diverses conditions physiques et hydrodynamiques à l'intérieur des pores. Par ailleurs, les mécanismes conduisant à l'adhésion de nano particules et de particules macromoléculaires sous conditions de non-équilibre, en raison de l'influence contradictoire des collisions mécaniques diffusifs et les forces attractives de Hamaker aux parois solides, sont d'un intérêt majeur. Nous avons donc développé un modèle théorique pour calculer le coefficient de restitution. L'objectif est de quantifier le bilan énergétique pendant le processus de collision diffusive de ces particules aux parois, sous l'influence des forces de répulsion d'une part et les forces attractives de Hamaker. Cela se fait par l'élaboration d'un modèle, basé sur le JKR et les théories d'Hertz, pour tenir compte des pertes d'énergie lors des collisions et des gains d'énergie en raison des interactions Hamaker. L'adhésion arrive si le bilan énergétique le permet. Notre modèle théorique est développé en proposant une approche particulière basée sur le potentiel Hamaker. Nous démontrons ce bilan par le biais d'une équation caractéristique non linéaire pour le coefficient de restitution, et analysons ses propriétés qui déterminent l'adhésion ou non pour diverses conditions physiques initiales.

<p>We investigate the use of noise degraded facial images in the application of facial expression recognition. In particular, we trained Gabor+SVMclassifiers to recognize facial expressions images with various types of noise. We applied Gaussian noise, Poisson noise, varying levels of salt and pepper noise, and speckle noise to noiseless facial images. Classifiers were trained with images without noise and then tested on the images with noise. Next, the classifiers were trained using images with noise, and then on tested both images that had noise, and images that were noiseless. Finally, classifiers were tested on images while increasing the levels of salt and pepper in the test set. Our results reflected distinct degradation of recognition accuracy. We also discovered that certain types of noise, particularly Gaussian and Poisson noise, boost recognition rates to levels greater than would be achieved by normal, noiseless images. We attribute this effect to the Gaussian envelope component of Gabor filters being sympathetic to Gaussian-like noise, which is similar in variance to that of the Gabor filters. Finally, using linear regression, we mapped a mathematical model to this degradation and used it to suggest how recognition rates would degrade further should more noise be added to the images.</p>

The primary objective of this study is to evaluate the seismic hazard in the Eastern Marmara Region using improved seismic source models and enhanced ground motion prediction models by probabilistic approach. Geometry of the fault zones (length, width, dip angle, segmentation points etc.) is determined by the help of available fault maps and traced source lines on the satellite images. State of the art rupture model proposed by USGS Working Group in 2002 is applied to the source system. Composite reoccurrence model is used for all seismic sources in the region to represent the characteristic behavior of North Anatolian Fault. New and improved global ground motion models (NGA models) are used to model the ground motion variability for this study. Previous studies, in general, used regional models or older ground motion prediction models which were updated by their developers during the NGA project. New NGA models were improved in terms of additional prediction parameters (such as depth of the source, basin effects, site dependent standard deviations, etc.), statistical approach, and very well constrained global database. The use of NGA models reduced the epistemic uncertainty in the total hazard incorporated by regional or older models using smaller datasets. The results of the study is presented in terms of hazard curves, deaggregation of the hazard and uniform hazard spectrum for six main locations in the region (Adapazari, Duzce, Golcuk, Izmit, Iznik, and Sapanca City Centers) to provide basis for seismic design of special structures in the area. Hazard maps of the region for rock site conditions at the accepted levels of risk by Turkish Earthquake Code (TEC-2007) are provided to allow the user perform site-specific hazard assessment for local site conditions and develop site-specific design spectrum. Comparison of TEC-2007 design spectrum with the uniform hazard spectrum developed for selected locations is also presented for future reference.

Traditionally, machine learning algorithms assume that training data is provided as a set of independent instances, each of which can be described as a feature vector. In contrast, many domains of interest are inherently multi-relational, consisting of entities connected by a rich set of relations. For example, the participants in a social network are linked by friendships, collaborations, and shared interests. Likewise, the users of a search engine are related by searches for similar items and clicks to shared sites. The ability to model and reason about such relations is essential not only because better predictive accuracy is achieved by exploiting this additional information, but also because frequently the goal is to predict whether a set of entities are related in a particular way. This thesis falls within the area of Statistical Relational Learning (SRL), which combines ideas from two traditions within artificial intelligence, first-order logic and probabilistic graphical models to address the challenge of learning from multi-relational data. We build on one particular SRL model, Markov logic networks (MLNs), which consist of a set of weighted first-order-logic formulae and provide a principled way of defining a probability distribution over possible worlds. We develop algorithms for learning of MLN structure both from scratch and by transferring a previously learned model, as well as an application of MLNs to the problem of Web query disambiguation. The ideas we present are unified by two main themes: the need to deal with limited training data and the use of bottom-up learning techniques. Structure learning, the task of automatically acquiring a set of dependencies among the relations in the domain, is a central problem in SRL. We introduce BUSL, an algorithm for learning MLN structure from scratch that proceeds in a more bottom-up fashion, breaking away from the tradition of top-down learning typical in SRL. Our approach first constructs a novel data structure called a Markov network template that is used to restrict the search space for clauses. Our experiments in three relational domains demonstrate that BUSL dramatically reduces the search space for clauses and attains a significantly higher accuracy than a structure learner that follows a top-down approach. Accurate and efficient structure learning can also be achieved by transferring a model obtained in a source domain related to the current target domain of interest. We view transfer as a revision task and present an algorithm that diagnoses a source MLN to determine which of its parts transfer directly to the target domain and which need to be updated. This analysis focuses the search for revisions on the incorrect portions of the source structure, thus speeding up learning. Transfer learning is particularly important when target-domain data is limited, such as when data on only a few individuals is available from domains with hundreds of entities connected by a variety of relations. We also address this challenging case and develop a general transfer learning approach that makes effective use of such limited target data in several social network domains. Finally, we develop an application of MLNs to the problem of Web query disambiguation in a more privacy-aware setting where the only information available about a user is that captured in a short search session of 5-6 previous queries on average. This setting contrasts with previous work that typically assumes the availability of long user-specific search histories. To compensate for the scarcity of user-specific information, our approach exploits the relations between users, search terms, and URLs. We demonstrate the effectiveness of our approach in the presence of noise and show that it outperforms several natural baselines on a large data set collected from the MSN search engine. / text