Prediction systems for College Entrance Examination (CEE) are popular during the graduating season, July every year in Taiwan. These systems give students suggestion according to their examination scores. There are several CEE prediction systems in Taiwan, but most of them are not constructed with rigorous theories. In 2005, Zen et al. constructed a prediction model using statistical method, which was later verified and improved by Lin in 2008. In this thesis, we will introduce the recording mechanism of the College Entrance Examination, and explain how to construct a prediction system under this mechanism. Also, we will compare the previous system with ours. We apply an empirical method and SVR as our first two approaches, and then we propose a new method. In our experiments, we consider the scores published by CEE center from 2004 to 2008. We use the root mean square error (RMSE) value to evaluate the performance of our present method. We also use the value generated by our method to show some information of the schools and the departments.

Ingeniero Civil Mecánico / El presente trabajo de título tiene como objetivo realizar una caracterización del estado de degradación de una bomba centrifuga en base a datos de emisiones acústicas (EA) medidas durante el período de operación hasta la falla. Las EA se pueden considerar como un indicador indirecto del daño, ya que permiten tener una noción de la evolución de este aún cuando no es directamente observable/medible. La distancia de Mahalanobis (DM), calculada a partir de las señales de EA obtenidas, permite obtener la medida de desviación de nuevas observaciones respecto a un conjunto de observaciones que den cuenta de un estado inicial. Con esto, es posible generar un índice de degradación a lo largo de la vida de operación del componente, tomando como el subconjunto de comparación a las mediciones que representan el estado saludable (sin degradación) del equipo bajo estudio. El diagnóstico de la bomba se realiza por medio de un filtro de partículas (FP), utilizado como método de inferencia dentro de una red Bayesiana dinámica (RBD). Esta permite representar la dependencia temporal y funcional entre todas las variables involucradas en el proceso de degradación considerado. Es necesario especificar cada dependencia dentro de la RBD. En particular se debe determinar el modelo de estado, que da cuenta de la evolución del daño en el tiempo, y el modelo de medición, que establece la relación entre las mediciones de EA con la degradación. En este caso, no existen modelos físicos que relacione las variables antes mencionadas, por lo tanto, ambos modelos se generan en base a datos. El modelo de estado es obtenido de una regresión polinomial entre los valores de la DM en base a la eficiencia de la bomba y el tiempo respectivo de cada medición. Para la generación del modelo de medición, se emplea la técnica de Support Vector Regression (SVR), la cual permite establecer una correlación no lineal entre las EA con el estado de daño. El FP emplea 1000 partículas para realizar la estimación del daño en cada instante de tiempo, este logra generar una estimación del daño de la bomba muy cercana a los valores de degradación real en el tiempo. Entre otras métricas de error, se obtuvo un coeficiente de determinación de R^2=0.9975. En base a los resultados, se puede concluir que el FP utilizado, en conjunto con los modelos generados, conducen un buen diagnóstico del estado de degradación de la bomba. Permitiendo tener una idea de la evolución del daño sufrido por la máquina a lo largo de su vida útil.

Orientador: Rejane Maria Tommasini Grotto / Banca: Dennis Armando Bertolini / Banca: Giovanni Faria Silva / Resumo: O vírus da Hepatite C (VHC) é um importante patógeno associado com doença hepática crônica sendo que alguns infectados podem desenvolver cirrose e carcinoma hepatocelular. O tratamento da hepatite C crônica visa a resposta virológica sustentada (RVS), definida como níveis de RNA viral indetectáveis no soro por seis meses depois do término do tratamento. Atualmente, a terapia padrão ouro é a combinação de interferon α peguilado e ribavirina, porém esse esquema terapêutico vem se mostrando eficaz em, apenas, 50% dos pacientes infectados com o genótipo 1, o mais prevalente no Brasil. Portanto, novas drogas mais eficazes e menos tóxicas estão sendo desenvolvidas para melhorar a assistência aos pacientes infectados pelo VHC, entre as quais merecem destaque os inibidores da serina protease NS3, a qual é uma enzima essencial para a replicação do VHC e assim um potencial alvo para novas terapias antivirais. Entretanto, a emergência de variantes resistentes é o maior obstáculo para o sucesso da terapêutica. Variantes resistentes já foram isoladas em pacientes tratados com os inibidores de protease e, estão associadas com a falência terapêutica. Porém o impacto dessas variantes resistentes em pacientes virgens de tratamento ainda não foi esclarecido e, esse tipo de informação pode avaliar o impacto dos inibidores de protease na terapia antiviral. O objetivo deste estudo foi avaliar a presença de mutações de resistência e polimorfismos genéticos na região genômica NS3 do VHC em 37 pacientes virgens de tratamento com inibidores de protease infectados com genótipo 1. RNA viral sérico foi utilizado como fonte para amplificação e seqüenciamento da região NS3 do VHC e, avaliar a presença de mutações de... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: The Hepatitis C Virus (HCV) is an important pathogen associated with chronic hepatic disease and some infected patients can develop cirrhosis and hepatocellular carcinoma. The treatment of chronic hepatitis C aimed the sustained virological response (SVR), defined as having undetectable serum HCV RNA at the end of therapy for at least 6 months. Currently, the gold standard therapy is a combination of pegylated interferon-α and the ribavirin, however this treatment present efficacy in only 50% of patients infected with genotypes 1, the most prevalent in Brazil. Then, new drugs more effective and less toxic have been developed to improve the attendance of the HCV infected patients as the serine protease NS3 inhibitors, which is an enzyme essential to HCV replication and main target of new antiviral therapies. However, the emergence of drug resistant variants has been the major obstacle to therapeutic successful. Resistant variants have already been isolated in patients treated with protease inhibitors and, these resistant variants are associated with non response to treatment. But the impact of the resistant variants in naïve protease inhibitors patients is unclear yet and, this information can evaluate the impact of protease inhibitors in antiviral therapeutic. The goal of this study was evaluate the presence of resistance mutations and genetic polymorphisms in the NS3 genomic region of HCV in 37 protease inhibitors-naive genotype 1 HCV infected patients. Serum viral RNA was used as source to amplification and sequencing of NS3 region of HCV and, evaluates the presence of resistance mutations and polymorphisms in this region. The results showed that only 07 (18.9%) samples presented resistant variants, the mutations... (Complete abstract click electronic access below) / Mestre

In recent years, the use of machine learning has increased significantly. Its uses range from making the everyday life easier with voice-guided smart devices to image recognition, or predicting the stock market. Predicting economic values has long been possible by using methods other than machine learning, such as statistical algorithms. These algorithms and machine learning models use time series, which is a set of data points observed constantly over a given time interval, in order to predict data points beyond the original time series. But which of these methods gives the best results? The overall purpose of this project is to predict Sweden’s aid curve using the machine learning model Support Vector Regression and the classic statistical algorithm autoregressive integrated moving average which is abbreviated ARIMA. The time series used in the prediction are annual summaries of Sweden’s total aid to the world from openaid.se since 1998 and up to 2019. SVR and ARIMA are implemented in python with the help of the Scikit- and Statsmodels libraries. The results from SVR and ARIMA are measured in comparison with the original value and their predicted values, while the accuracy is measured in Root Square Mean Error and presented in the results chapter. The result shows that SVR with the RBF-kernel is the algorithm that provides the best results for the data series. All predictions beyond the times series are then visually presented on a openaid prototype page using D3.js / Under det senaste åren har användningen av maskininlärning ökat markant. Dess användningsområden varierar mellan allt från att göra vardagen lättare med röststyrda smarta enheter till bildigenkänning eller att förutspå börsvärden. Att förutspå ekonomiska värden har länge varit möjligt med hjälp av andra metoder än maskininlärning, såsom exempel statistiska algoritmer. Dessa algoritmer och maskininlärningsmodeller använder tidsserier, vilket är en samling datapunkter observerade konstant över en given tidsintervall, för att kunna förutspå datapunkter bortom den originella tidsserien. Men vilken av dessa metoder ger bäst resultat? Projektets övergripande syfte är att förutse sveriges biståndskurva med hjälp av maskininlärningsmodellen Support Vector Regression och den klassiska statistiska algoritmen autoregressive integrated moving average som förkortas ARIMA. Tidsserien som används vid förutsägelsen är årliga summeringar av biståndet från openaid.se sedan år 1998 och fram till 2019. SVR och ARIMA implementeras i python med hjälp av Scikit-learn och Statsmodelsbiblioteken. Resultatet från SVR och ARIMA mäts i jämförelse mellan det originala värdet och deras förutspådda värden medan noggrannheten mäts i root square mean error och presenteras under resultatkapitlet. Resultatet visar att SVR med RBF kärnan är den algoritm som ger det bästa testresultatet för dataserien. Alla förutsägelser bortom tidsserien presenteras därefter visuellt på en openaid prototypsida med hjälp av D3.js.

In this thesis, least squares support vector machine for regression (LS-SVR) is applied to image coding. First, we propose five simple algorithms for solving LS-SVR. For linear regression, two simple Widrow-Hoff-like algorithms, in primal form and in dual form, are proposed for LS-SVR problems. The dual form of the algorithm is then generalized to kernel-based nonlinear LS-SVR. The elegant and powerful two-parameter sequential minimization optimization (2PSMO) and three-parameter sequential minimization optimization (3PSMO) algorithms are provided in detail. A predictive function obtained from LS-SVR is utilized to approximate the gray levels of the image. After pruning, only a subset of training data called support vectors is saved. Experimental results on seven image blocks show that the LS-SVR with Gaussian kernel is more appropriate than that with Mahalanobis kernel with a covariance matrix. Two-layer LS-SVR is proposed to choose the machine parameters of the LS-SVR. Before training outer LS-SVR, feature extraction is used to reduce the input dimensionality. Experimental results on three whole images show that the results with two-layer LS-SVR after reducing dimensionality are better than those with two-layer LS-SVR without reducing dimensionality in PSNR for Lena and Baboon images and they are almost the same in PSNR for F16 image.

[pt] Esta dissertação tem por objetivo investigar métodos de estimação da idade a partir de imagens faciais. Avalia-se o impacto de distintos fatores sobre a acurácia da estimativa, especificamente, a acurácia da localização de pontos fiduciais, métodos de extração de atributos, de redução de dimensionalidade, e técnicas de regressão. Adicionalmente, foi estudada a influência da raça e do sexo na acurácia da estimação da idade desenvolvido. Consideraram-se cinco métricas de desempenho do sistema, especificamente, o erro médio absoluto (MAE), o erro médio absoluto por década (MAE/D), o erro médio absoluto por idade (MAE/A), o escore acumulado (CS), e os intervalos de confiança (IC). Os experimentos foram realizados empregando dois bancos de dados públicos, cujas imagens estão rotuladas com a idade da face. Os resultados indicaram que o método automático para detecção de pontos fiduciais da face tem uma repercussão moderada sobre a acurácia das estimativas. Entre as variantes analisadas, a que apresentou a melhor acurácia foi o sistema que emprega os AAMs (Active Appearance Models) como método de extração de atributos, o PCA (Principal Components Analysis) como método para reduzir dimensionalidade, e as SVRs (Support Vector Regression) como técnica para fazer regressão. / [en] This thesis aims to investigate methods for age estimation from facial images. The impact of distinct factors over the estimate’s accuracy is assessed, specifically the accuracy in the location of face fiducial points, feature extraction and dimensionality reduction methods, and regression techniques. Additionally, the dependence on race and gender in the accuracy of age estimation is assessed. Five performance metrics have been considered: the mean absolute error (MAE), the mean absolute error per decade (MAE / D), the mean absolute error for age (MAE / A), the cumulative score (CS) and confidence intervals (CI). The experiments were performed using two public databases, whose images are labeled with the age of the face. The results showed the impact of the automatic method for detection of fiducial points of the face has a moderate impact on the accuracy of the estimates. Among the analyzed variants, the one with the best accuracy was the system that employs the Active Appearance Models (AAMs) as feature extraction method, the Principal Components Analysis (PCA) as dimensionality reduction method, and Support Vector Regression (SVRs) as a technique to do regression.

Die DOR-Plastik (Surgical Ventricular Reconstruction, SVR) ist ein chirurgisches Verfahren zur Rekonstruktion ventrikulärer kardialer Strukturen bei Herzinsuffizienz-Patienten mit apikaler A- und Dyskinesie. Jedoch gibt es spätestens seit dem negativen Ergebnis einer großen multizentrischen Studie (STICH-trial, Jones et al. 2009 [1]) eine Kontroverse über den tatsächlichen prognostischen Nutzen der Operation. Ziel dieser Arbeit war es in diesem Zusammenhang mittels kardialer Magnet-Resonanz-Tomographie (Cardiac Magnetic Resonance, CMR) generierte potentielle Prädiktoren der funktionellen Erholung nach der DOR-Plastik zu analysieren. Dafür wurden in dieser Arbeit bei 24 Patienten die kardialen Volumina, die kardiale Funktion, das lokale und totale myokardiale Narbengewebe und verschiedene geometrische Indizes bestimmt und ausgewertet. Es konnte gezeigt werden, dass die quantitative Ermittlung des basalen myokardialen Narbengewebes und des apikalen Volumenindex (AVI) dabei helfen könnten, eine Subgruppe von Patienten zu definieren, die von der DOR-Plastik profitiert.

Made available in DSpace on 2014-06-11T19:23:08Z (GMT). No. of bitstreams: 0 Previous issue date: 2011-02-15Bitstream added on 2014-06-13T19:28:48Z : No. of bitstreams: 1 zeminian_lb_me_botfm.pdf: 506409 bytes, checksum: 08bea6131bbfbe54e0327803c5f6bb14 (MD5) / Ministério da Saúde / O vírus da Hepatite C (VHC) é um importante patógeno associado com doença hepática crônica sendo que alguns infectados podem desenvolver cirrose e carcinoma hepatocelular. O tratamento da hepatite C crônica visa a resposta virológica sustentada (RVS), definida como níveis de RNA viral indetectáveis no soro por seis meses depois do término do tratamento. Atualmente, a terapia padrão ouro é a combinação de interferon α peguilado e ribavirina, porém esse esquema terapêutico vem se mostrando eficaz em, apenas, 50% dos pacientes infectados com o genótipo 1, o mais prevalente no Brasil. Portanto, novas drogas mais eficazes e menos tóxicas estão sendo desenvolvidas para melhorar a assistência aos pacientes infectados pelo VHC, entre as quais merecem destaque os inibidores da serina protease NS3, a qual é uma enzima essencial para a replicação do VHC e assim um potencial alvo para novas terapias antivirais. Entretanto, a emergência de variantes resistentes é o maior obstáculo para o sucesso da terapêutica. Variantes resistentes já foram isoladas em pacientes tratados com os inibidores de protease e, estão associadas com a falência terapêutica. Porém o impacto dessas variantes resistentes em pacientes virgens de tratamento ainda não foi esclarecido e, esse tipo de informação pode avaliar o impacto dos inibidores de protease na terapia antiviral. O objetivo deste estudo foi avaliar a presença de mutações de resistência e polimorfismos genéticos na região genômica NS3 do VHC em 37 pacientes virgens de tratamento com inibidores de protease infectados com genótipo 1. RNA viral sérico foi utilizado como fonte para amplificação e seqüenciamento da região NS3 do VHC e, avaliar a presença de mutações de... / The Hepatitis C Virus (HCV) is an important pathogen associated with chronic hepatic disease and some infected patients can develop cirrhosis and hepatocellular carcinoma. The treatment of chronic hepatitis C aimed the sustained virological response (SVR), defined as having undetectable serum HCV RNA at the end of therapy for at least 6 months. Currently, the gold standard therapy is a combination of pegylated interferon-α and the ribavirin, however this treatment present efficacy in only 50% of patients infected with genotypes 1, the most prevalent in Brazil. Then, new drugs more effective and less toxic have been developed to improve the attendance of the HCV infected patients as the serine protease NS3 inhibitors, which is an enzyme essential to HCV replication and main target of new antiviral therapies. However, the emergence of drug resistant variants has been the major obstacle to therapeutic successful. Resistant variants have already been isolated in patients treated with protease inhibitors and, these resistant variants are associated with non response to treatment. But the impact of the resistant variants in naïve protease inhibitors patients is unclear yet and, this information can evaluate the impact of protease inhibitors in antiviral therapeutic. The goal of this study was evaluate the presence of resistance mutations and genetic polymorphisms in the NS3 genomic region of HCV in 37 protease inhibitors-naive genotype 1 HCV infected patients. Serum viral RNA was used as source to amplification and sequencing of NS3 region of HCV and, evaluates the presence of resistance mutations and polymorphisms in this region. The results showed that only 07 (18.9%) samples presented resistant variants, the mutations... (Complete abstract click electronic access below)

Chemical substances are essential to all aspects of human life, and understanding their properties is essential for effective application. The properties of chemical species are usually measured by experimentation or computational calculation using theoretical methods. In this work, machine learning models (ML) for predicting entropy, S, and heat capacity, cp, were developed for alkanes, alkenes, and alkynes at 298.15 K. The data for entropy and heat capacity were collected from various sources. Commercial software (alvaDesc) then generated the molecular descriptors of all the hydrocarbons in the dataset used as input for the ML models. Support vector regression (SVR), v-support vector regression (v-SVR), and random forest regression (RFR) algorithms were trained with K-fold cross-validation on two levels. The first level assessed the models’ performance and the second level generated the final models. After a performance comparison of the three models, the SVR was chosen. To illustrate the advantage of using the ML approach, the SVR model was compared against Benson’s group additivity. Finally, a sensitivity analysis was performed.

Cette thèse a pour but de modéliser les liquides ioniques et les électrolytes pour batteries Li-ion. Nous avons développé des modèles SVR afin de prédire 9 propriétés d’intérêt pour ces solvants. Les modèles construits pour les liquides ioniques ont permis la détection de divers problèmes, et sont accessibles sur le site web du laboratoire : infochim.u-strasbg.fr/webserv/VSEngine.html. Les modèles construits pour les électrolytes ont permis la modélisation de candidats testés expérimentalement par nos collaborateurs. Le nombre de données étant limité pour ces solvants, nous avons également testé l’approche transductive par le biais de la TRR (Transductive Ridge Regression). Nous avons mis en place un protocole d’optimisation des paramètres de la méthode et appliqué la TRR aux solvants étudiés. Les résultats obtenus par la TRR sont légèrement meilleurs que ceux de la Régression Ridge, mais restent modestes si on veut éviter une détérioration accidentelle du modèle. / This thesis is dedicated to the modelling of ionic liquids and electrolytes of Li-ion batteries. We developed several SVR models in order to predict 9 interesting properties of these solvents. The models built for the ionic liquids allowed us to detect several problems, and are freely available on the laboratory’s website: infochim.u-strasbg.fr/webserv/VSEngine.html. The models built for the electrolytes were used to model some candidates tested experimentally by our colleagues. As the amount of data is quite small for these solvents, we also tested the transductive approach with the help of the TRR (Transductive Ridge Regression). We have developed an optimization procedure for the method’s parameters, and applied the TRR to the studied solvents. The results obtained with the TRR are slightly better than of the Ridge Regression but stay modest if we want to avoid any accidental damage of the model.