Over the past years early detection and intervention in schizophrenia have become a major objective in psychiatry. Early intervention strategies are intended to identify and treat psychosis prior to fulfilling diagnostic criteria for the disorder. To this aim, reliable early diagnostic biomarkers are needed in order to identify a high-risk state for psychosis and also predict transition to frank psychosis in those high-risk individuals destined to develop the disorder. Recently, machine learning methods have been successfully applied in the diagnostic classification of schizophrenia and in predicting transition to psychosis at an individual level based on magnetic resonance imaging (MRI) data and also neurocognitive variables. This work investigates the application of machine learning methods for the early identification of schizophrenia in subjects at high risk for developing the disorder. The dataset used in this work involves data from the Edinburgh High Risk Study (EHRS), which examined individuals at a heightened risk for developing schizophrenia for familial reasons, and the FePsy (Fruherkennung von Psychosen) study that was conducted in Basel and involves subjects at a clinical high-risk state for psychosis. The overriding aim of this thesis was to use machine learning, and specifically Support Vector Machine (SVM), in order to identify predictors of transition to psychosis in high-risk individuals, using baseline structural MRI data. There are three aims pertaining to this main one. (i) Firstly, our aim was to examine the feasibility of distinguishing at baseline those individuals who later developed schizophrenia from those who did not, yet had psychotic symptoms using SVM and baseline data from the EHRS study. (ii) Secondly, we intended to examine if our classification approach could generalize to clinical high-risk cohorts, using neuroanatomical data from the FePsy study. (iii) In a more exploratory context, we have also examined the diagnostic performance of our classifier by pooling the two datasets together. With regards to the first aim, our findings suggest that the early prediction of schizophrenia is feasible using a MRI-based linear SVM classifier operating at the single-subject level. Additionally, we have shown that the combination of baseline neuroanatomical data with measures of neurocognitive functioning and schizotypal cognition can improve predictive performance. The application of our pattern classification approach to baseline structural MRI data from the FePsy study highly replicated our previous findings. Our classification method identified spatially distributed networks that discriminate at baseline between subjects that later developed schizophrenia and other related psychoses and those that did not. Finally, a preliminary classification analysis using pooled datasets from the EHRS and the FePsy study supports the existence of a neuroanatomical pattern that differentiates between groups of high-risk subjects that develop psychosis against those who do not across research sites and despite any between-sites differences. Taken together, our findings suggest that machine learning is capable of distinguishing between cohorts of high risk subjects that later convert to psychosis and those that do not based on patterns of structural abnormalities that are present before disease onset. Our findings have some clinical implications in that machine learning-based approaches could advise or complement clinical decision-making in early intervention strategies in schizophrenia and related psychoses. Future work will be, however, required to tackle issues of reproducibility of early diagnostic biomarkers across research sites, where different assessment criteria and imaging equipment and protocols are used. In addition, future projects may also examine the diagnostic and prognostic value of multimodal neuroimaging data, possibly combined with other clinical, neurocognitive, genetic information.

Strength of concrete is the major parameter in the design of structures and is represented by the 28-day compressive strength of concrete. Many earlier studies proved that the compressive strength of concrete is not only related to w/c ratio but also rely on proportion of other constituent materials. Application of recently developed new generation admixtures for the production of high performance concrete, has made the concrete strength prediction complex and highly nonlinear challenging the research engineers and data scientists. Development of early accurate prediction model for concrete strength provides the mix designer a tentative idea to proportionate the mix ingredients accordingly reducing the number of trial mixes ultimately saving a lot of cost and time associated with it. In this study, we have proposed SVM regression tool to create the model for the prediction of concrete strength. Support vector machine (SVM) is a supervised machine learning technique based on statistical learning theory developed by Vapnik in 1995. SVM employs a kernel function to transform the data into high dimensional feature space and linear modeling is performed in the feature space to overcome the complexity related to highly nonlinear datasets. A dataset containing 425 observations of high performance concrete mix design with nine attribute variables from University of California, Irvine Repository are considered for this study. 395 datasets were used to train the model and 30 samples were taken as a test set by random sub sampling to test the model. Five-fold cross-validation technique was used to select the parameters of SVM. The metaparameter values ε = 0.001, C = 29.47 and γ = 10 are selected for creating the model. The model performance measures correlation coefficient (R), root mean square error (RMSE) values and residual plots suggest that the proposed SVM model is competent enough to predict the strength of concrete. The performance measures of proposed SVM model was compared with RVM model.

Early and accurate prediction of the compressive strength of concrete is important in the construction industry. Modeling the compressive strength of concrete to obtain a balance and equality between prediction accuracy, time and uncertainty of the prediction is a very difficult task due to the highly nonlinear nature of concrete. For structural engineering purposes, the 28- day compressive strength is the most relevant parameter. In this study, an attempt has been made to predict the 28-day compressive strength of concrete using Relevance Vector Machine (RVM). An RVM belongs to the class of sparse kernel classifiers, which are powerful tools in classification and regression. It has a model of identical functional form to the popular and state-of-the-art `Support Vector Machine (SVM)'. The benefits of using RVM include automatic estimation of nuisance parameters, probabilistic prediction and the ability to model complex data with little information. A total of 425 different data of high performance mix designs were collected from the University of California, Irvine repository. The data used to predict the compressive strength consisted of nine components. The RVM model was trained and tested using 395 and 30 data sets respectively. The model's performance was assessed at the end of the training and testing period using four performance measures; coefficient of determination, root-mean-square error, percentage of relevance vectors and residual plots. All the performance measures confirmed the accuracy of the model. The results of the study suggested that RVM is an effective tool for predicting the 28- day compressive strength of concrete from its mix ingredients.

Diseases can have a range of impacts on hosts and host populations. These impacts can be minimal, to the point of being considered nearly a commensal relationship. The other end of the spectrum is when a disease regulates a population or even drives it to extinction. Diseases that are directly transmitted in a density-dependent manner typically do not cause population extinction because as the population decreases so does transmission. However, there are several factors that can lead to extinction caused by diseases. These diseases can be frequency-dependent transmission (including vector-borne diseases) or diseases that infect multiple sympatric hosts. The parasite Cytauxzoon felis is a tick-borne apicomplexan that infects bobcats and domestic cats and is enzootic in the Midwest and Southeast US. Bobcats are considered the reservoir host of C. felis and typically do not show signs of disease associated with infection. This parasite is the etiological agent of cytauxzoonosis, a highly-fatal disease impacting domestic cats and occasionally other felid species. Domestic cats also can be subclinically infected. This parasite has increased its range and density within previously described enzootic areas. There are several aspects of the biology of C. felis that have not been explored and impact the epizootiology of this important veterinary parasite. The prevalence of the parasite has been studied in some locations, yet areas where the parasite has recently invaded need to be studied as populations of naïve bobcats and domestic cats may be at risk of epizootics. My research expands the knowledge of C. felis, adding information about a relatively recently-described enzootic area. The present dissertation is divided into 6 chapters. In the first chapter I provide a thorough literature review of C. felis and general information on pathogens. In my second chapter I describe the prevalence of C. felis in bobcats and ticks in southern Illinois from 2006-2017. This is the first documentation of C. felis in bobcats in Illinois. The prevalence in ticks is also the highest prevalence in ticks reported to date. These results have been published in the Journal of Parasitology. In the third chapter I provide evidence of chronic C. felis infections in bobcats. Some bobcats maintained C. felis infection for at least 2 years. I determined individual bobcats were infected with the same strain of the parasite at each capture event. This finding indicates that bobcats may carry the same strain over time. These infected bobcats could be spreading strains that are more pathogenic to domestic cats and possibly to other bobcats. Vector-borne pathogens (specifically microparasites) can show varying levels of intensity of infection in vertebrate hosts. The intensity of infection may correlate with activity of the vectors to facilitate transmission from the vertebrate to the vector. In the fourth chapter I tested if C. felis parasitemia (percent of red blood cells infected with the parasites) increased with environmental factors associated with tick activity. Cytauxzoon felis infections are increasing in domestic cats in the US in many areas. The fifth chapter describes the first study of clinical and subclinical C. felis infections in domestic cats in southern Illinois. I collaborated with veterinary clinics to obtain 642 domestic cat blood samples for this project. I also tested whether land cover types and host characteristics were related to risk of infection, and found only that feral cats were more likely to have subclinical C. felis infection. Cytauxzoon felis is transmitted through a tick vector; so direct contact between domestic cats and bobcats is not necessary for transmission to occur. For the sixth chapter I tested if the genetic populations of C. felis in domestic cats and bobcats were different (suggesting barriers to transmission). I found that there was high genetic diversity of C. felis in my samples. The within population variance accounted for nearly all variance detected. Therefore I conclude that the population of C. felis in bobcats and domestic cats in my study area is panmictic suggesting there are no barriers to transmission between these two host species.

El presente trabajo de tesis propone una metodología para la construcción de un modelo predictivo de fugas voluntarias, el cual tiene como objetivo la identificación temprana de los clientes que presentan mayores tendencias a la fuga.

Made available in DSpace on 2016-08-29T15:33:12Z (GMT). No. of bitstreams: 1 tese_3521_.pdf: 2782962 bytes, checksum: d4b2294e5ee9ab86b7a35aec083af692 (MD5) Previous issue date: 2010-02-12 / Os Modelos de Predição implementados pelos algoritmos de Aprendizagem de Máquina advindos como linha de pesquisa da Inteligência Computacional são resultantes de pesquisas e investigações empíricas em dados do mundo real. Neste contexto; estes modelos são extraídos para comparação de duas grandes técnicas de aprendizagem de máquina Redes Neuro-Fuzzy e de Regressão aplicadas no intuito de estimar um parâmetro de qualidade do produto em um ambiente industrial sob processo contínuo. Heuristicamente; esses Modelos de Predição são aplicados e comparados em um mesmo ambiente de simulação com intuito de mensurar os níveis de adequação dos mesmos, o poder de desempenho e generalização dos dados empíricos que compõem este cenário (ambiente industrial de mineração).

Made available in DSpace on 2014-06-11T19:32:18Z (GMT). No. of bitstreams: 0 Previous issue date: 2006-11-28Bitstream added on 2014-06-13T19:21:51Z : No. of bitstreams: 1 simao_fb_dr_rcla.pdf: 1851483 bytes, checksum: 4f667b0db3f8c30e9f84faf0b4bb1ca0 (MD5) / Unir/Riomar / O município de Porto Velho teve sua ocupação marcada por um sério desordenamento urbano, que resultou, entre outros, em aglomerados populacionais de baixa renda, originados de migrações externas e internas que se assentaram em locais impróprios para a urbanização. Isso deu origem a conflitos sócio-ambientais e de saúde pública, ou seja, contribuiu para uma crescente insuficiência dos serviços de saneamento e para o incremento da pobreza. A deterioração das condições de vida no município criou ambientes favoráveis à proliferação de vetores transmissores de doenças parasitárias, contribuindo para a intensificação da transmissão no meio urbano e para o agravamento do problema de saúde pública, em especial pela malária. A maioria da população portovelhense habita áreas com elevado risco de transmissão, pois vive em locais com ambientes propícios à formação de criadouros dos mosquitos que transmitem a malária, tais como: nascentes, drenagens, áreas alagadas e florestas remanescentes. Há também, em algumas áreas, evidência de vegetação equatorial úmida, de maneira a tornar a cidade vulnerável à proliferação de mosquitos anofelinos, vetores de malária, especialmente em áreas com freqüência de migrantes. O município, hoje, detém pouco mais de 25% dos casos de malária do Estado de Rondônia, isso porque o crescimento populacional acelerado cria uma série de problemas urbanos de infra-estrututa em todos os aspectos. O modelo de políticas públicas atuais não é eficiente na erradicação da doença na área urbana. Em razão dos problemas expostos, esse estudo teve por objetivo mapear as possíveis áreas de maior prevalência do vetor transmissor da doença, associando-as aos problemas sócio-ambientais para que essas informações possam orientar o poder público e a sociedade, com o fim de interferir no controle da endemia eficientemente. / Porto Velho municipal district had its occupation characterized by a serious disordering process, which has resulted in agglomerations of low income people, originated from external and internal migration which had been settled in improper areas, therefore this occupation originated social-environmental and public health conflicts, contributing to an increasing sewage inadequacy and poverty. The degradation process of life standards in the municipal district led to favorable conditions for the proliferation of vector transmitter parasitic diseases, contributing to the disease spreading in urban areas and the aggravation of public health problems, particularly malaria. The majority of Porto Velho's population inhabits areas where there are several transmission risks, due the high incidence of malaria transmitter mosquitoes that live in water nascent, drainage, flooded and forest remained areas. There is also, in some areas, equatorial humid vegetation, turning the place vulnerable to the proliferation of malaria mosquito transmitters, particularly in areas with high frequency of migrants. The municipal district has nowadays more than 25% of malaria cases in Rondonia, mainly because of disordered population increasing, which creates urban problems in all aspects. The public politics used nowadays aren't efficient on eradicating the disease in urban areas. Due to the exposed problems, this study aims to map the potential malaria prevalence areas, associating to the socio-environment problems in order to orientate government and society to control, properly, the endemic malaria. In order to accomplish this task, there was used specific statistical techniques of multiple correspondence analysis, applied to physical environment and adapted to the social area.

abstract: Gas turbine engine for aircraft propulsion represents one of the most physics-complex and safety-critical systems in the world. Its failure diagnostic is challenging due to the complexity of the model system, difficulty involved in practical testing and the infeasibility of creating homogeneous diagnostic performance evaluation criteria for the diverse engine makes. NASA has designed and publicized a standard benchmark problem for propulsion engine gas path diagnostic that enables comparisons among different engine diagnostic approaches. Some traditional model-based approaches and novel purely data-driven approaches such as machine learning, have been applied to this problem. This study focuses on a different machine learning approach to the diagnostic problem. Some most common machine learning techniques, such as support vector machine, multi-layer perceptron, and self-organizing map are used to help gain insight into the different engine failure modes from the perspective of big data. They are organically integrated to achieve good performance based on a good understanding of the complex dataset. The study presents a new hierarchical machine learning structure to enhance classification accuracy in NASA's engine diagnostic benchmark problem. The designed hierarchical structure produces an average diagnostic accuracy of 73.6%, which outperforms comparable studies that were most recently published. / Dissertation/Thesis / Masters Thesis Electrical Engineering 2015

The purpose of this essay is to answer the following questions: Has the increased e-commerce had a negative impact on the inflation, and is the effect decreasing? and: Is there a long term and/or short term effect by the increased e-commerce on the inflation? To answer the first question a fixed effects regression model is applied, based on panel data for 28 European countries for the time period 2006-2017. The regression obtains results that support the hypothesis that the increased e-commerce has had a negative effect on inflation. Furthermore, the result indicates that the effect is decreasing. The second question is answered with the help of an Error Correction Model and time series data for Sweden during the period 2006-2017. The result shows that there is an error correction towards a long run equilibrium and the short term estimates indicate that there is a negative short term effect of the increased e-commerce on inflation. These results are in line with the hypothesis of this essay as well as previous studies that have examined similar questions.

Submitted by FABIANA DA SILVA FRANÇA (fabiana21franca@gmail.com) on 2017-11-09T15:38:24Z No. of bitstreams: 2 arquivototal.pdf: 3033279 bytes, checksum: f2b56b36fa23f100495d8d218c7e3669 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Made available in DSpace on 2017-11-09T15:38:24Z (GMT). No. of bitstreams: 2 arquivototal.pdf: 3033279 bytes, checksum: f2b56b36fa23f100495d8d218c7e3669 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Previous issue date: 2013-08-30 / Dengue is a viral systemic disease caused by an arboviral of Flaviviridae family, affecting about a 100 million cases per year in Brazil. It is endemic in tropical regions such as Southeast Asia, South Pacific, East Africa, Caribbean and Latin America. The disease is transmitted by Aedes aegypti (Linnaeus, 1762), a mosquito that is the main target for the disease control through strategies ranging from the larval to the adult combat. The larvicides commonly used to combat the vector, besides being toxic, present drop in larvicide efficacy since the A. aegypti larvae has developed resistance to these products. Thus, the search for new active principles that are effective in combating the mosquito is required. In this sense, Agave sisalana is a plant that is produced in several states in the Brazilian northeast region, which is used in the sisal industry. Only 5% of the plant is recovered, and its residual liquid completely wasted. In this way, the aim of this research project was to investigate the larvicidal action of the juice of Agave sisalana against larvae of A. aegypti. In larvicidal activity assays, fourth stage A. aegypti larvae were used, exposed to different concentrations of A. sisalana liquid waste during 24 hours. After the larvicidal activity assays, it was possible to determine the LC50 that was 5.9 mg / mL. Next we explored the cytotoxic activity of A. sisalana in hemocytes of A. aegypti larvae through the flow cytometry. The experiments showed an increase of cellular necrosis after 12 hours of exposure of the larvae to submaximal concentrations of sisal liquid waste (7.4% in control group vs. 28.5% in the experimental group after 12 hours; 6.2% in the control group vs. 22.7% in the experimental group after 24 hours). The histological alterations were confirmed by histopathological analysis, which showed lyses of the mesentery epithelial cells of larvae as well as peritrophic membrane destruction. Furthermore, nitric oxide (NO) production by hemocytes, an important defense strategy of mosquitoes, was checked after 3, 6 and 24 hours of larvae exposure to the A. sisalana liquid waste. There was a reduction in NO levels of approximately 76.6% after 3 hours, 83% after 6 hours and 83.8% after 24 hours of exposure. In this way, the A. sisalana liquid waste constitutes an effective alternative and economically feasible for the dengue vector combat. The outcomes of our research resulted in the patent application for an insecticide against A. aegypti larvae. / A dengue é uma doença viral sistêmica, causada por um arbovírus da família Flaviviridae, acometendo cerca de 700 mil casos por ano no Brasil. É endêmica de regiões tropicais como o sudeste asiático, sul do Pacífico, África Oriental, Caribe e América Latina. A dengue é transmitida pelo mosquito Aedes aegypti (Linnaeus, 1762), que é o principal alvo de combate para controle da doença, por meio de estratégias que vão desde o combate às formas larvares até o mosquito adulto. Os larvicidas comumente utilizados no combate do vetor, além de serem tóxicos, vêm apresentando queda na capacidade larvicida já que as larvas do A. aegypti tem desenvolvido resistência a esses produtos. Sendo assim, a busca por novos princípios ativos que sejam eficientes no combate do mosquito se faz necessária. Nesse sentido, a Agave sisalana é uma planta que é produzida em vários estados do nordeste brasileiro, a qual é utilizada na indústria sisaleira. Apenas 5% da planta é aproveitada, sendo o seu resíduo líquido completamente desperdiçado. Dessa forma, este projeto de pesquisa teve como objetivo investigar a ação larvicida do suco de Agave sisalana contra larvas de A. aegypti. Nos ensaios de atividade larvicida, utilizou-se larvas de quarto estágio de A. aegypti, testando-se diferentes concentrações de suco de A. sisalana durante 24 horas. Após os ensaios de atividade larvicida foi possível determinar a CL50, que foi de 5,9 mg/mL. A pesquisa também explorou a atividade citotóxica da A. sisalana em hemócitos de larvas de A. aegypti, através da citometria de fluxo. Verificou-se um aumento no percentual de necrose celular a partir de 12 horas de exposição das larvas a concentrações submáximas de suco de sisal (7,4% no grupo controle vs. 28,5% no grupo experimental após 12 horas; 6,2% no grupo controle vs.22,7% no grupo experimental após 24 horas). As alterações histológicas foram confirmadas em exames histopatológicos, que mostraram lise celular de células epiteliais do mesentério das larvas e destruição da membrana peritrófica. A produção de óxido nítrico (NO) pelos hemócitos, uma importante estratégia de defesa dos mosquitos, foi verificada após 3,6 e 24 horas de exposição das larvas ao suco de A. sisalana. Observou-se uma diminuição dos níveis de NO da ordem de 76,6% após 3 horas de exposição, 83 % após 6 horas de exposição, e 83,8 % após 24 horas de exposição. Sendo assim, o suco de A. sisalana pode se constituir numa alternativa efetiva e economicamente viável para o combate ao vetor da Dengue. Essa pesquisa resultou no pedido de patente de um inseticida formulação a base de A. sisalana para combate às larvas de A. aegypti.