Bayesian networks can be used to model the respiratory system. Their structure indicate how risk factors, symptoms, and diseases are related and the Conditional Probability Tables enable predictions about a patient’s need for hospitalization. Numerous structure learning algorithms exist for discerning the structure of a Bayesian network, but none can guarantee to find the perfect structure. Employing multiple algorithms can discover relationships between variables that might otherwise remain hidden when relying on a single algorithm. The Maximum Likelihood Estimator is the predominant algorithm for learning the Conditional Probability Tables. However, it faces challenges due to the data fragmentation problem, which can compromise its predictions. Failing to hospitalize patients who require specialized medical care could lead to severe consequences. Therefore, in this thesis, the use of an XGBoost model for learning is proposed as a novel and better method since it does not suffer from data fragmentation. A Bayesian network is constructed combining several structure learning algorithms, and the predictive performance of the Maximum Likelihood Estimator and XGBoost are compared. XGBoost achieved a maximum accuracy of 86.0% compared to the Maximum Likelihood Estimator, which attained an accuracy of 81.5% in predicting future patient hospitalization. In this way, the predictive performance of Bayesian networks has been enhanced. / Bayesianska nätverk kan användas för att modellera andningssystemet. Deras struktur visar hur riskfaktorer, symtom och sjukdomar är relaterade, och de villkorliga sannolikhetstabellerna möjliggör prognoser om en patients behov av sjukhusvård. Det finns många strukturlärningsalgoritmer för att urskilja strukturen i ett bayesianskt nätverk, men ingen kan garantera att hitta den perfekta strukturen. Genom att använda flera algoritmer kan man upptäcka relationer mellan variabler som annars kan förbli dolda när man bara förlitar sig på en enda algoritm. Maximum Likelihood Estimator är den dominerande algoritmen för att lära sig de villkorliga sannolikhetstabellerna. Men den står inför utmaningar på grund av datafragmenteringsproblemet, vilket kan äventyra dess prognoser. Att inte lägga in patienter som behöver specialiserad medicinsk vård kan leda till allvarliga konsekvenser. Därför föreslås i denna avhandling användningen av en XGBoost-modell för inlärning som en ny och bättre metod eftersom den inte lider av datafragmentering. Ett bayesianskt nätverk byggs genom att kombinera flera strukturlärningsalgoritmer, och den prediktiva prestandan för Maximum Likelihood Estimator och XGBoost jämförs. XGBoost uppnådde en maximal noggrannhet på 86,0% jämfört med Maximum Likelihood Estimator, som uppnådde en noggrannhet på 81,5% för att förutsäga framtida patientinläggning. På detta sätt har den prediktiva prestandan för bayesianska nätverk förbättrats.

A conceptual design of an innovative two-axle lightweight railway vehicle for commuter services is carried out at KTH Railway Group. An active wheelset steering is introduced to improve the curving performance of the vehicle, which is one of the critical performance requirements. This thesis aims to improve the steering performance of the active wheelset steering. Look-up tables for estimating time-varying wheel-rail contact parameters are introduced to supervise a simple PID controller of the active steering system in order to improve steering performance. The look-up table (LUT) estimation is focused on time-varying wheel-rail contact parameters, including creep coefficients and contact patch variables due to their direct influence on curving performance and lateral stability of the wheelset. As a result, the estimated longitudinal unit creep forces (UCF) have the potential to supervise the gains determination of PID controller because it can appropriately distinguish running conditions. The estimation of longitudinal UCF is achieved by the combination of the results from the LUT of creep coefficients and the LUT of contact patch variables. The result from longitudinal unit creep force estimation is shifted to the first quadrant to use as critical gain in the Ziegler-Nichols tuning method for the PID controller. The critical oscillation period for PID tuning can be expressed as a function of vehicle speed. Consequently, the PID controller for the active steering system uses time-varying gains with real-time tuning. The proposed control system for active wheelset steering is validated with nine running conditions using SIMPACK and MATLAB/Simulink co-simulation. The proposed control system provides a stable wheelset lateral displacement control regardless of the running condition. The active steering system significantly reduces wheel-rail wear, which demonstrates the effectiveness of the proposed active steering system. / KTH:s Järnvägsgruppen utvecklar en konceptuell design av ett innovativt, två-axligt, lättvikts järnvägsfordon för tunnelbana eller pendeltåg. En aktiv hjuparsstyrning introduceras för att förbättra kurvtagningsförmågan hos fordonet, vilket är ett av de kritiska prestandakraven hos dessa fordon. Det här examensarbetet har som målsättning att förbättra styrningsprestandan av den aktiva hjulsatsstyrningen. För att uppskatta tidsvarierande hjul-rälskontaktparametrar introduceras pre-definierade tabeller (LUT) som en övervakning av en enkel PID-kontroll för det aktiva styrningssystemet, för att förbättra styrprestandan. Uppskattningen som baseras på tabellen fokuserar på tidsberoende hjul-rälsparametrar, inklusive krypkoefficienter och kontaktytans storlek och form. Dessa variabler är i fokus på grund av deras direkta effekt på kurvtagningsförmågan och den laterala stabiliteten hos hjulparet. Den uppskattade longitudinala enhets krypkraften (UCF) har potential att bestämma förstärkningen hos PID-kontrollen på grund av att den, på ett lämpligt sätt, kan skilja mellan olika körtillstånd. Uppskattningen av longitudinell UCF uppnås genom en kombination av resultat för krypkoefficienter och kontaktytavariabler i LUT. Resultaten från den longitudinella UCF-uppskattningen skiftas till den första kvadranten för att användas som kritisk förstärkning i Ziegler-Nichols justeringsmetod för PID-kontroller. Den kritiska oscillationsperioden för PID-justering kan utryckas som en funktion av fordonets hastighet. Utgående från detta använder PID-kontrollen tidsvarierande förstärkning med realtidsjustering för den aktiva styrningen. Det föreslagna kontrollsystemet valideras mot nio körtillstånd med hjälp av SIMPACK och MATLAB/Simulink-simuleringar. Det föreslagna kontrollsystemet tillhandahåller en stabil lateral förflyttning av hjulparet oberoende av körtillstånd. Det aktiva styrsystemet reducerar hjul-räls slitaget signifikant, vilket demonstrerar effektiviteten hos det framtagna aktiva styrsystemet.

Cette étude a pour objectif la description syntaxique et sémantique des constructions transitives non locatives à un complément d'objet direct en grec moderne : N0 V N1. Nous nous sommes appuyée sur le cadre théorique de la grammaire transformationnelle de Zellig S. Harris et sur le cadre méthodologique du Lexique-Grammaire, défini par Maurice Gross et développé au Laboratoire d'Automatique Documentaire et Linguistique. À partir de 16 560 entrées verbales morphologiques, nous procédons à la classification des constructions transitives non locatives, à partir de 24 classes distinctes, sur la base de critères formels posés. Un inventaire de 2 934 emplois verbaux à construction transitive non locative à un complément d'objet direct a été ainsi produit et scindé en neuf classes. Parmi ces emplois, 1 884 sont formellement décrits dans 9 tables de lexique-grammaire établies : plus précisément, il s'agit de celles qui impliquent des constructions à un complément d'objet direct illustrant les concepts « apparition » (table 32GA), « disparition » (32GD), objet « concret » (32GC), « partie du corps » (32GCL), substantif « humain » (32GH), substantif avec « pluriel obligatoire » (32GPL). En outre, la transformation passive est largement interdite pour les emplois verbaux recensés dans la table 32GNM, alors que les tables 32GCV et 32GRA regroupent des verbes acceptant une transformation à verbe support. Nous présentons l'application des données linguistiques recensées dans le traitement automatique des langues naturelles (TALN), avec la conversion automatique des tables en automates à états finis récursifs, suivie de nos suggestions sur leur applicabilité à la traduction en français et à l'enseignement du grec moderne (langue maternelle ou étrangère) : acquisition/apprentissage / The current research aims to provide a syntactic and semantic analysis of Modern Greek transitive non-locative constructions with one direct object: N0 V N1. Our study is based on the syntactic framework of the Transformational Grammar defined by Zellig S. Harris. We followed the Lexicon-Grammar methodology framework developed by Maurice Gross and elaborated at the LADL (Laboratoire d'Automatique Documentaire et Linguistique). Based on 16 560 morphological verbal entries, we proceeded to the classification of transitive non-locative constructions. On the basis of formal criteria we divided them into 24 distinct classes that formed an inventory of 2 934 transitive non-locative verbal uses with one direct object. Among them, 1 884 verbal uses were split into nine classes and they were formally described in 9 lexicon-grammar tables established for this purpose. More precisely, these structures include a direct object referring to the following concepts: “appearance” (32GA table), “disappearance” (32GD), “concrete” object (32GC), “body part” (32GCL), “human” object (32GH), and “obligatory plural” (32GPL). Likewise, the passive transformation is largely blocked in the 32GNM table, while the 32GCV and 32GRA tables regroup verbs accepting a support verb transformation. We present the linguistic data application in Natural Language Processing (NLP), by means of automatic tables conversion into recursive transition network automata. Moreover, we set forth our remarks on their applicability in translation from Modern Greek to French as well as in language learning/teaching (Modern Greek as first or second language)

Membrane proteins represent about one third of all known vertebrate proteins and over half of the current drug targets. Knowledge of their three-dimensional (3D) structure is worth millions of pounds to the pharmaceutical industry. Yet experimental structure elucidation of membrane proteins is a slow and expensive process. In the absence of experimental data, computational modelling tools can be used to close the gap between the numbers of known protein sequences and structures. However, currently available structure prediction tools were developed with globular soluble proteins in mind and perform poorly on membrane proteins. This thesis describes the development of a modelling approach able to predict accurately the structure of transmembrane domains of proteins. In this thesis we build a template-based modelling framework especially for membrane proteins, which uses membrane protein-specific information to inform the modelling process.Firstly, we develop a tool to accurately determine a given membrane protein structure's orientation within the membrane. We offer an analysis of the preferred substitution patterns within the membrane, as opposed to non-membrane environments, and how these differences influence the structures observed. This information is then used to build a set of tools that produce better sequence alignments of membrane proteins, compared to previously available methods, as well as more accurate predictions of their 3D structures. Each chapter describes one new piece of software or information and uses the tools and knowledge described in previous chapters to build up to a complete accurate model of a transmembrane domain.

Selected methods of mortality analysis focused on adults and the oldest age-groups Abstract Questions about human life span, longevity and mortality in general are natural to almost everyone. This Doctoral Thesis deals with one central question - whether some limit of human life span or of its improvements exists. It is rather a methodological work, therefore its aim is to introduce not only relevant theories but above all the methods usable in the mortality analysis focused on adults or the oldest-old. At the beginning the most important theories and opinions of scientist dealing with mortality are introduced. In the first half of the analytical part mainly the traditional and basic approaches are included. The theme of life tables is opened by an analysis of its construction in the Czech Republic, together with its possible modifications. As a result the independent macro code for the SAS software is attached in the electronic Appendix. This macro enables to calculate the unknown parameters of selected mortality laws by the method of weighted non-linear least squares and to produce the smoothed and extrapolated values of mortality rates. Using the individual life durations, life tables according to education attainment were constructed (also attached in the electronic Appendix). In the second half of the...

[pt] Os custos de instalação e as taxas de produção ao longo da vida de um reservatório de óleo e gás são influenciados diretamente pela localização, quantidade e capacidade das Unidades Estacionárias de Produção (UEPs). A distância entre um poço e a UEP a qual foi alocado é um fator impactante na perda de carga a que os fluídos são submetidos. A dissipação de energia aumenta quando essa distância é maior e todo o sistema de produção recebe a interferência negativa desta perda, o que compromete as taxas de recuperação. A necessidade de respeitar as restrições de capacidade das UEPs faz com que outras decisões precisem ser tomadas no mesmo momento em que se decide a localização de cada uma. Este trabalho descreve um modelo baseado em Algoritmos Genéticos para a otimização simultânea da quantidade, localização e dimensionamento de Unidades Estacionárias de Produção (UEPs). Para lidar com as restrições lineares e não lineares do problema utiliza-se a técnica chamada de GENOCOP III - Genetic Algorithm for Numerical Optimization of Constrained Problems e funções de penalidade. O objetivo da otimização é maximizar o Valor Presente Líquido (VPL) que depende da curva de produção de cada configuração obtida como possível solução. Para obter a curva de produção são realizadas simulações de reservatório que utilizam tabelas de escoamento multifásico para representar o sistema de produção externo ao reservatório. O modelo de solução foi testado em um modelo de reservatório baseado em um caso real. Os resultados encontrados indicam que a utilização deste modelo de solução como ferramenta pode auxiliar a tomada de decisão dos especialistas responsáveis pelo desenvolvimento de campos de petróleo. / [en] Installation costs and production rates over the life of an oil and gas reservoir are directly influenced by the location, number and capacity of the Production Units. The distance between a well and the Production Unit to which it has been allocated is an important factor in the loss of fluids pressure. The power dissipation increases when the distance is bigger and the entire production system receives the negative interference of this loss, compromising recovery rates. There is a need to take into account restrictions that apply to the capacity of Production Unit at the same time as there localization are decided. This paper describes a model with genetic algorithms for the simultaneous optimization of the quantity, location and sizing of Production Units. To deal with the constraints of the problem we use a technique called GENOCOP III - Genetic Algorithm for Numerical Optimization of Constrained Problems. The goal of the optimization is to maximize the Net Present Value (NPV) which depends on the production curve of each configuration obtained as a possible solution. The production curves are obtained by reservoir simulations with multiphase flow tables that represent the system external to the reservoir. The solution model was tested in a reservoir model based on a real case. The results indicate that using this solution model as a tool can assist the decision making of experts responsible for oil field development.

Infant mortality in the Czech Republic and Europe: trends and patterns Abstract This Master's thesis addresses the development of infant mortality in the Czech Republic between the years 1950 and 2016 using more detailed indicators of infant mortality. To evaluate the impact of infant mortality rate, on the lengthening life expectancy at birth in the Czech Republic and selected European countries, a decomposition of life tables is included. Selected European countries are then compared with the use of cluster analysis, based on infant mortality indicators and life expectancy at birth by sex. Finally, an analysis is made on the basis of infant life tables which are calculated by sex, birth weight and legitimacy for the Czech Republic. Keywords: infant mortality, Czech Republic, European countries, tables of infant mortality, sex, child-legitimacy, birth weight, decomposition, cluster analysis

Made available in DSpace on 2015-12-03T18:33:12Z (GMT). No. of bitstreams: 1 Henry_Marcondes_Santos_Lopes.pdf: 1552777 bytes, checksum: 8c2cafcae0b53d4373d90165f6c26d80 (MD5) Previous issue date: 2014-11-04 / This paper analyzes the impact of changes in the Brazilian economy s basic interest rate, the Selic, on the result of active insurers in Brazil within dec/2002 and dec/2013. Through panel data regressions, it was observed how the profitability of different companies on the insurance market behaved due to variations in the interest rate. The objective is to point out the practices with which the insurance company try to optimize the returns on their operations, including making them less dependent on the outcome of the financial account. The insurers accounting data used in the paper were extracted from the Superintendence of Private Insurance (SUSEP) statistical system, and the results show that variations in the Selic rate had statistically significant impacts on the financial result and the premium charged by activeinsurance companies in Brazil, and the companies apparently offset lower profitability of the financial account by increasing the premiums charged, which may be an evidence of oligopolistic behavior of this market in Brazil. / O trabalho analisa o impacto de modifica????es na taxa b??sica de juros da economia brasileira, a SELIC, no resultado das seguradoras atuantes no Brasil no per??odo de dez/2002 a dez/2013. Por meio de regress??es utilizando dados em painel, verificou-se como a rentabilidade de diferentes empresas do ramo de seguros se comportou em decorr??ncia das varia????es ocorridas na taxa de juros. Procura-se assim apontar as pr??ticas com as quais as sociedade seguradoras tentam otimizar os retornos de suas opera????es, inclusive tornando-os menos dependentes dos resultados da conta financeira. No estudo foram utilizados dados cont??beis das seguradoras extra??dos do sistema de estat??sticas da Superintend??ncia de Seguros Privados (SUSEP) e os resultados encontrados evidenciam que varia????es da SELIC tiveram impactos estatisticamente significantes no resultado financeiro e no pr??mio cobrado pelas Seguradoras atuantes no Brasil, sendo que as empresas aparentemente compensaram uma menor rentabilidade da conta financeira com acr??scimos nos pr??mios cobrados, o que pode evidenciar um comportamento oligopolista deste mercado no Brasil.

Made available in DSpace on 2016-04-27T17:12:58Z (GMT). No. of bitstreams: 1 Leticia de Castro Araujo.pdf: 1350079 bytes, checksum: e53936b39991f19e6a9fa7bcc7bd3029 (MD5) Previous issue date: 2007-05-09 / Made available in DSpace on 2016-08-25T17:25:38Z (GMT). No. of bitstreams: 2 Leticia de Castro Araujo.pdf.jpg: 2104 bytes, checksum: c4715912a635b5fbde63d2a9b070733f (MD5) Leticia de Castro Araujo.pdf: 1350079 bytes, checksum: e53936b39991f19e6a9fa7bcc7bd3029 (MD5) Previous issue date: 2007-05-09 / Secretaria da Educação do Estado de São Paulo / I summarize starting from 1997 the Statistics teaching was included by the Parameters National Curriculares as one of the blocks of mathematical content to be taught in the initial series of the elementary school. To they analyze the grating curricular of some courses of Pedagogy; some researchers alerted for the scarce time in that the disciplines related to the Statististics are worked. In that way, the teacher s qualification is committed for the appearance of gaps in his formation that they impede him of understanding his role in the construction of that knowledge. However, the teacher needs to face some challenges to help the students in the construction of the knowledge in relation to the Statistics. The tables and graphs, mainly with respect to the extrapolation starting of the tendency of data. Another great challenge concerns the introduction of the concepts as average, modal value, medium, in such a way that these are not restricted to the use of the algorithm to calculate them, but that are to worked with meaning starting from situations that make sense for the students and justify their calculations. The objective of this work is the starting from a test diagnosis, to investigate which are the conceptions and competences of a group of educator relates to the reading and interpretation of tables and graphs. The test was divided in objective and dissertativas, and in the first category we will analyze the teacher s conception regarding the theme, when locating characteristics starting from numeric data in tables and graphs, as well as, if x necessary, to formalize calculations: and in second, the competences, that is, which strategies used by him that you will help him to justify their answers. To we wake into account the results obtained in our research, we conclude that the formation of those teaches raising for experiences shared by the friends which is not linked necessarily to academic banks, but it plays fundamental part in the conception and the teacher s competence on learning and teaching of the Mathematics / A partir de 1997 o Tratamento da Informação foi incluído pelos Parâmetros Curriculares Nacionais como um dos blocos de conteúdo matemático a ser ensinado nas séries iniciais do Ensino Fundamental. Ao analisarem as grades curriculares de alguns cursos de Pedagogia, alguns pesquisadores alertaram para o tempo escasso em que as disciplinas relacionadas à Estatística são trabalhadas. Dessa forma, a qualificação do Professor polivalente ou pedagogo fica comprometida pelo surgimento de lacunas em sua formação que o impedem de compreender o seu papel na construção desse conhecimento. Entretanto, o professor precisa enfrentar alguns desafios para ajudar os alunos na construção do conhecimento em relação à Estatística. O primeiro deles diz respeito à leitura e interpretação de tabelas e gráficos, principalmente no que tange à extrapolação a partir da tendência de dados. Outro grande desafio diz respeito à introdução dos conceitos como média moda e mediana, de tal forma que estes não fiquem restritos ao uso do algoritmo para calculá-los, mas que sejam trabalhados com significado a partir de situações que façam sentido para os alunos e justifiquem seus cálculos. O objetivo deste trabalho é a partir de um teste diagnóstico, investigar quais são as concepções e competências de um grupo de professores polivalentes relacionadas à leitura e interpretação de tabelas e gráficos. O teste foi dividido em questões objetivas e dissertativas, e na primeira categoria iremos analisar a concepção do professor a respeito do tema, ao localizar características a partir de dados numéricos em tabelas e gráficos, bem como se necessário, formalizar cálculos: e na segunda, as competências, viii isto é, quais estratégias utilizadas por ele que o ajudarão a justificar suas respostas. Ao levarmos em conta os resultados obtidos em nossa pesquisa, concluímos que a formação desses professores passa por experiências compartilhadas pelos colegas a qual não necessariamente está vinculada a bancos acadêmicos, mas desempenha papel fundamental na concepção e competência do professor sobre aprendizagem e ensino da Matemática

Changes in the marital status in the age group from 15 to 59 years in the Czech Republic after 1930 Abstract The aim of this thesis is to analyze the changes in population structure according to the individual marital statuses. For this analysis sex-specific and age-specific nuptuality rates and multistate life tables were use, in order to be able to describe transitions between the individual marital statuses. In the first part of this thesis the data sources and the methods of creating the transversal increment- decrement life tables were described. Following chapters deal with the historical development of the Czech Republic population since the year 1930 and the trends of nuptuality, divorce and mortality. Second part is dedicated to the analysis of the transitions between the individual marital statuses. The findings of this thesis conclude that people in the individual marital states tend to react similarly to the historical events in Czech Republic, unless some events are purposefully aiming to influence only a part of the population. Key words: nuptuality rate, divorce rate, mortality rate, marital status, multistate demography, transversal increment-decrement life tables, Czech Republic