This thesis explores the evolution plural verbal -s ("People thinks he is guilty") and zero 3rd singular ("He think he is guilty") in data from two sources on Southern English: The Linguistic Atlas of the Gulf States (LAGS) and The Linguistic Atlas of the Middle and South Atlantic States (LAMSAS). The research questions that underlie this study consider (1) the demographic association of plural verbal -s and zero 3rd singular, (2) the maintenance of each form, (3) the constraints on their use, and (4) the origins of -s variability. The atlas data suggest the following for plural verbal -s: (1) it has a British source, (2) it was present in both African American Vernacular English (AAVE) and early Southern White English (SWE), and (3) there were different grammatical constraints on its use in AAVE and SWE. Data for zero 3rd singular -s suggest this form (1) did not have a British source and (2) that it has historically been an AAVE feature.

Um potencial de dois corpos do tipo delta-Dirac foi utilizado para descrever sistemas fracamente ligados de três corpos. A trajetória completa dos estados Efimov em função da energia ligação de dois corpos foi calculada para o caso de três bosons idênticos: se o subsistema de dois corpos é ligado, conforme a razão entre a energia de ligação de dois e três corpos aumenta, o estado excitado desaparece e um estado virtual correspondente aparece quando a energia do estado fundamental atinge o limiar dado por 6.9hbar^2/(ma^2) (a - comprimento de espalhamento, m - massa do bóson). Quando o subsistema de dois corpos é virtual, o aumento da razão entre as energias faz com que o estado excitado se transforme em uma ressonância quando a energia do estado fundamental é 1.1hbar^2/(ma^2). Neste último caso as condições para a formação de moléculas triatômicas no interior de condensados é favorecida, pois a competição com dímeros fracamente ligados está ausente. A energia de ligação de trímeros com momento angular total nulo em condensados atômicos foi estimada através da correlação desta com o coeficiente de recombinação e com a energia do dímero, ambos conhecidos experimentalmente em alguns casos. Os tamanhos de moléculas fracamente ligadas (^4He_2$-X; Xequiv^4He, ^6Li, ^7Li e ^{23}Na) e de núcleos exóticos leves (^6He, ^{11}Li, ^{14}Be e ^{20}C) também foram calculados juntamente com um estudo sistemático do comportamento dos raios quadráticos médios conforme a interação dos subsistemas de dois corpos é variada. Neste último caso a classificação de um sistema de três corpos (tipo AAB) foi completada denominando de Samba a configuração formada por dois subsistemas de dois corpos ligados e um virtual. As equações subtraídas para os estados ligados de quatro bosons também foram deduzidas.

Este trabalho apresenta o estudo de um carregador de baterias monofÃsico aplicado a veÃculos elÃtricos. Este carregador à composto por dois estÃgios de processamento de energia e um circuito digital de supervisÃo para controlar a tensÃo sobre o banco de baterias e a corrente de recarga das mesmas. O primeiro estÃgio consiste de um conversor CA-CC bridgeless com caracterÃstica de alto fator de potÃncia, e o segundo estÃgio à representado por um conversor CC-CC fullbridge com isolamento em alta frequÃncia e comutaÃÃo sob tensÃo nula (Zero Voltage Switching â ZVS). Para ambos os conversores, foi realizada uma anÃlise qualitativa e quantitativa, bem como apresentados exemplos de projeto para facilitar o dimensionamento dos componentes. Finalmente, com os componentes escolhidos, foi montado um protÃtipo que permite carregar de uma atà oito baterias de 12 V conectadas em sÃrie. O sistema apresenta como especificaÃÃes: tensÃo de entrada alternada de 220 VÂ15%; tensÃo de saÃda contÃnua de 120 V; corrente de saÃda contÃnua de 20 A; e potÃncia mÃdia de saÃda de 2,4 kW. / This work presents a single-phase battery charger for electric vehicles. This converter is composed by two energy processing stages and a digital circuit to control the voltage across the batteries and their respective charging current. The first stage is a high power factor ACDC bridgeless converter, while the second one consists on a ZVS (Zero Voltage Switching) high frequency isolated DC-DC full-bridge converter. For both converters, the qualitative and quantitative analyses have been performed, as well as design examples have been presented in order to ease the components calculation. Finally, a prototype that allows charging up to eight series-connected 12 V batteries has been built. The system specifications are: AC input voltage of 220 V Â15%; DC output voltage of 120 V; DC output current of 20 A; and average output power of 2.4 kW.

Submitted by Sandro Camargo (sandro.camargo@unipampa.edu.br) on 2015-05-09T22:33:43Z No. of bitstreams: 1 126110043.pdf: 2117805 bytes, checksum: f107c883cc20c40fff355188fc0e7df6 (MD5) / Made available in DSpace on 2015-05-09T22:33:44Z (GMT). No. of bitstreams: 1 126110043.pdf: 2117805 bytes, checksum: f107c883cc20c40fff355188fc0e7df6 (MD5) Previous issue date: 2014-08-20 / Atualmente as técnicas não lineares de modulação de potência aplicadas ao aquecimento resistivo de água, principalmente em redes elétricas mal projetadas, podem gerar vários problemas no ambiente onde estão instaladas. É o caso dos chuveiros eletrônicos que, dependendo da técnica utilizada para executar o processo de modulação de potência, podem ocasionar efeitos indesejáveis na rede elétrica e ambiente. Um exemplo típico é o Controle por Ciclos Inteiros, foco deste trabalho, que apresenta baixa probabilidade de geração de harmônicas e interferência eletromagnética (IEM). Em contrapartida, é praticamente inevitável o surgimento de flicker na rede elétrica decorrente de sua utilização, principalmente devido às altas potências dos equipamentos de aquecimento resistivo, agravadas pelos longos intervalos entre condução e não condução de corrente. Considerando que a geração de flicker é a principal desvantagem proveniente do Controle por Ciclos Inteiros, o presente trabalho tem por objetivo propor outras técnicas a fim de atenuar este efeito indesejável e, ao mesmo tempo, possibilitar a utilização de vários níveis de potência. / Currently the non-linear power modulation techniques applied to resistive water heating, mainly on not properly designed electric installations, can generate various problems in the environment where they are present. This is the case of electronic showers that depending on the power modulation technique may result in undesirable effects in the power grid and environment. A typical example is the zero-crossing AC power control, focus of this work, which has low probability of generating harmonics and electromagnetic interference (EMI). On the other hand, it is almost inevitable the emergence of Flicker on the power grid resulting from its use, mainly due to high electrical power consumed by heating equipment, compounded by long conducting and non-conducting power intervals. Whereas the flicker generated is the main drawback from the zero-crossing power control, this work aims to propose other techniques to reduce this undesirable effect allowing several output power levels.

Coding Electronic Medical Records (EMRs) with diagnosis and procedure codes is an essential task for billing, secondary data analyses, and monitoring health trends. Both speed and accuracy of coding are critical. While coding errors could lead to more patient-side financial burden and misinterpretation of a patient’s well-being, timely coding is also needed to avoid backlogs and additional costs for the healthcare facility. Therefore, it is necessary to develop automated diagnosis and procedure code recommendation methods that can be used by professional medical coders. The main difficulty with developing automated EMR coding methods is the nature of the label space. The standardized vocabularies used for medical coding contain over 10 thousand codes. The label space is large, and the label distribution is extremely unbalanced - most codes occur very infrequently, with a few codes occurring several orders of magnitude more than others. A few codes never occur in training dataset at all. In this work, we present three methods to handle the large unbalanced label space. First, we study how to augment EMR training data with biomedical data (research articles indexed on PubMed) to improve the performance of standard neural networks for text classification. PubMed indexes more than 23 million citations. Many of the indexed articles contain relevant information about diagnosis and procedure codes. Therefore, we present a novel method of incorporating this unstructured data in PubMed using transfer learning. Second, we combine ideas from metric learning with recent advances in neural networks to form a novel neural architecture that better handles infrequent codes. And third, we present new methods to predict codes that have never appeared in the training dataset. Overall, our contributions constitute advances in neural multi-label text classification with potential consequences for improving EMR coding.

Public school students across the United States have been criminalized for minor youth behavior issues such as truancy, defiance, and minor fighting incidents. The presence of law enforcement is expanding in school spaces, increasing the likelihood of young students facing court systems for minor offenses. Criminalization of students is counterproductive considering schools are designed to promote student growth and development. Little is known about how students and parents experience school criminalization. The purpose of this multi-case study, based on Freire's conceptual framework of critical consciousness, was to investigate how a small group of families experienced school criminalization. Three families of youths who had been criminalized for minor school offences were recruited using community partners as referral sources. Interviews were conducted with parents using a semi-structured protocol, and data were also obtained from school and court records provided by parents. Data were triangulated, summarized as case descriptions, member checked, and then cross-theme analyzed based on Gibbs and Taylor's approach for emergent themes. Study results demonstrated that these families felt trapped between two institutions and experienced fear and frustration trying to deal with both systems. Participants also recommended ways parents and schools might improve discipline for minor offences. This study will influence social change by informing school and juvenile justice discipline policy reform about working with two systems in managing student behavior concerns. In addition, the interview protocol can be used by human services professionals to help improve understanding of clients faced with school criminalization issues.

This manuscript consists of two papers that formulate novel methodologies pertaining to time series analysis in the state-space framework. In Chapter 1, we introduce an innovative time series forecasting procedure that relies on model-based clustering and model averaging. The clustering algorithm employs a state-space model comprised of three latent structures: a long-term trend component; a seasonal component, to capture recurring global patterns; and an anomaly component, to reflect local perturbations. A two-step clustering algorithm is applied to identify series that are both globally and locally correlated, based on the corresponding smoothed latent structures. For each series in a particular cluster, a set of forecasting models is fit, using covariate series from the same cluster. To fully utilize the cluster information and to improve forecasting for a series of interest, multi-model averaging is employed. We illustrate the proposed technique in an application that involves a collection of monthly disease incidence series. In Chapter 2, to effectively characterize a count time series that arises from a zero-inflated binomial (ZIB) distribution, we propose two classes of statistical models: a class of observation-driven ZIB (ODZIB) models, and a class of parameter-driven ZIB (PDZIB) models. The ODZIB model is formulated in the partial likelihood framework. Common iterative algorithms (Newton-Raphson, Fisher Scoring, and Expectation Maximization) can be used to obtain the maximum partial likelihood estimators (MPLEs). The PDZIB model is formulated in the state-space framework. For parameter estimation, we devise a Monte Carlo Expectation Maximization (MCEM) algorithm, using particle methods to approximate the intractable conditional expectations in the E-step of the algorithm. We investigate the efficacy of the proposed methodology in a simulation study, and illustrate its utility in a practical application pertaining to disease coding.

On introduit un nouveau modele de file d'attente: les files Zero-automatiques. Tout d'abord, on considere la discipline de service Premier Arrive Premier Servi. Les files 0-automatiques sont caracterisees par une salle d'attente evoluant suivant un mecanisme de marche aleatoire sur un groupe ou un monoede infini. En considerant les deux cas les plus simples et aussi extremes de files 0-automatiques, nous retrouvons la file simple M/M/1 et la G-file de Gelenbe avec clients positifs et negatifs.<br />Le resultat saillant est que toutes les files 0-automatiques ont une distribution stationnaire a forme produit et un processus de depart de Poisson. C'est un point crucial pour construire les reseaux a forme produit.<br />On considere deux modeles correspondant aux differents routages classiques: reseaux a la Jackson et reseaux a la Kelly. Dans les deux cas, on a montre que la distribution stationnaire est a forme produit et peut etre determinee explicitement. De plus, le processus de depart est Poisson.<br />Enfin, considerons les files 0-automatiques avec discipline de service Dernier Arrive Premier Servi. Dans ce cas, certaines proprietes restent vraies, mais pas toutes. On obtient des resultats interessants en comparant les zones de stabilite d'une meme file 0-automatique sous les discpilines Premier Arrive Premier Servi et Dernier Arrive Premier Servi.

<p>Due to the invention of automated hacking tools, Hacking is not a black art anymore. Even script kiddies can launch attacks in few seconds. Therefore, there is a great emphasize on the Security to protect the resources from camouflage. Intrusion Detection System is also one weapon in the security arsenal. It is the process of monitoring and analyzing information sources in order to detect vicious traffic. With its unique capabilities like monitoring, analyzing, detecting and archiving, IDS assists the organizations to combat against threats, to have a snap-shot of the networks, and to conduct Forensic Analysis. Unfortunately there are myriad products inthe market. Selecting a right product at time is difficult. Due to the wide spread rumors and paranoia, in this work I have presented the state-of-the-art IDS technologies, assessed the products, and evaluated. I have also presented some of the novel challenges that IDS products are suffering. This work will be a great help for pursuing IDS technology and to deploy Intrusion Detection Systems in an organization. It also gives in-depth knowledge of the present IDS challenges.</p>

This thesis examines various signal processing techniques that are required for establishing efficient (near optimal) communications in multiuser multiple-input multiple-output (MIMO) environments. The central part of this thesis is dedicated to acquisition of information about the MIMO channel state - at both the receiver and the transmitter. This information is required to organize a communication set up which utilizes all the available channel resources. Realistic channel model, i.e., the spatial channel model (SCM), has been used in this study, together with modern long-term evolution (LTE) standard. The work consists of three major themes: (a) estimation of the channel at the receiver, also known as tracking; (b) quantization of the channel information and its feedback from receiver to the transmitter (feedback quantization); and (c) reconstruction of the channel knowledge at the transmitter, and its use for data precoding during communication transmission. / Communications