Submitted by Marlucy Farias Medeiros (marlucy.farias@unipampa.edu.br) on 2018-05-11T18:34:32Z No. of bitstreams: 1 Vinícius Bittencourt da Silva- 2018.pdf: 4217704 bytes, checksum: 062fdfd3aa89f2dbd2679f836c3bad2d (MD5) / Approved for entry into archive by Dayse Pestana (dayse.pestana@unipampa.edu.br) on 2018-05-14T11:58:52Z (GMT) No. of bitstreams: 1 Vinícius Bittencourt da Silva- 2018.pdf: 4217704 bytes, checksum: 062fdfd3aa89f2dbd2679f836c3bad2d (MD5) / Made available in DSpace on 2018-05-14T11:58:52Z (GMT). No. of bitstreams: 1 Vinícius Bittencourt da Silva- 2018.pdf: 4217704 bytes, checksum: 062fdfd3aa89f2dbd2679f836c3bad2d (MD5) Previous issue date: 2018-02-07 / Atualmente, o tempo de inserção de um produto de hardware no mercado é cada vez menor apesar do crescimento de sua complexidade. Portanto, é importante que o processo de construção seja cada vez mais rápido. Entre as medidas para ganhar desempenho a otimização do tempo despendido em verificação é fundamental, pois cerca de 70% do tempo de projeto é aplicado nessa atividade. Esse processo inicia-se juntamente com o desenvolvimento, pois, caso seja detectado um erro somente no estágio final de desenvolvimento é possível que haja atrasos para cumprir os prazos de entrega. Nesse sentido, este trabalho apresenta a USAG, uma ferramenta semi-automática desenvolvida para construir ambientes de verificação usando a metodologia UVM (a qual é a metodologia padrão atualmente) aplicada ao projeto de circuitos integrados escritos em SystemVerilog. Esta ferramenta vem para ajudar no processo de verificação de hardware acelerando a criação do ambiente de verificação, uma vez que ele gera as estruturas e interconexões da metodologia e produz os arquivos para simulação. Qualquer ferramenta que suporte SystemVerilog juntamente com a Metodologia UVM pode executar o ambiente de verificação gerado pela USAG. Além disso, a ferramenta é baseada na Web para ser acessível a partir de qualquer local sem a necessidade de um sistema operacional específico ou configuração para usá-la. Finalmente, são apresentados os resultados de ambientes de verificação UVM obtidos a partir da entrada de códigos fonte em SystemVerilog na USAG. A partir dos resultados obtidos e da análise da utilização por parte de testadores conclui-se que a USAG é eficaz no que tange os objetivos propostos. / Currently, the insertion time of a hardware pro ducts in the market is decreasing despite the growth of its complexity. Therefore, it is important that the construction process is getting faster and faster. Among the ways to gain performance, the optimization of the time spent in verification is fundamental, because ab out 70% of the project time is applied in this activity. This process starts with the development, because if an error is detected only in the final stage of development, there may be delays to comply with delivery times. In this way, this work presents USAG, a semi-automatic tool developed to construct verification environments using the UVM methodology (which is the current standard methodology) applied to the design of integrated circuits written in SystemVerilog. This tool comes to assist in the hardware verification process by accelerating the creation of the verification environment as it generates the structures and interconnections of the methodology and produces the files for simulation. Any tool that supports SystemVerilog together with the UVM Methodology can execute the verification environment generated by the USAG. In addition, the tool is web-based to be accessible from any lo cation without the need for a specific operating system or configuration to use it. Finally, the results of UVM verification environments obtained from the entry of source co des in SystemVerilog in USAG are presented.

The problematics of gigabit optical networks as well as web technologies suitable for a web tool implementation were studied within this thesis. An experimental web application for monitoring GPON frames is developed based on the proposed system architecture. The frontend is implemented using ReactJS and the Tornado web framework is used for backend implementation. Data for analysis are read from the stream using the Kafka platform.

Submitted by Nadir Basilio (nadirsb@uninove.br) on 2017-08-04T21:26:50Z No. of bitstreams: 1 Renato Alessandro Rocha Santos.PDF: 5054326 bytes, checksum: 739f1115bab9b157f48eb56ca44d4c34 (MD5) / Made available in DSpace on 2017-08-04T21:26:50Z (GMT). No. of bitstreams: 1 Renato Alessandro Rocha Santos.PDF: 5054326 bytes, checksum: 739f1115bab9b157f48eb56ca44d4c34 (MD5) Previous issue date: 2017-03-02 / Over the last decades the Vehicle Routing Problem (VRP) has been subject of research of several authors, mainly because of difficulties found in its optimization as well as its application in real-world situations. In the scientific literature there are several proposed solutions for the PRV using different optimization techniques. However, such solutions are rarely transformed into software tools that can be used by end users, for example, micro enterprises. Thus, the focus of the present work was to develop a Web computational tool for vehicle routing, called “SGRV  Sistema de Gestão de Roteamento de Veículos”, which uses Google Maps features and aims to meet the needs of micro enterprises. Therefore, initially it was conducted a literature search about solution methods for VRP, to support the choice of strategy based on genetic algorithms employed in SGRV. Then, a new bibliographic research was made with the purpose of finding free softwares for the VRP solution, which were object of analysis to mark out the development of the SGRV. For the development of this research was used the methodology Design Science Research. A qualitative evaluation of the SGRV was carried out by four microenterprises from different branches, that used the tool for a certain period and, in the end, answered six questions opened from semi-structured interviews. The experiences of these microenterprises were transcribed and the data obtained reveal the effectiveness of the SGRV in the management of its tasks related to the orders and deliveries of products. / Nas últimas décadas o Problema de Roteamento de Veículos (PRV) tem sido temática de pesquisas de diversos autores, principalmente por causa de dificuldades encontradas para sua resolução, bem como sua aplicabilidade em situações reais do cotidiano. Na literatura científica há diversas propostas de soluções para o PRV empregando diferentes técnicas de otimização. No entanto, tais soluções raramente são transformadas em ferramentas computacionais que possam ser utilizadas por usuários finais como, por exemplo, as microempresas. Assim, o foco do presente trabalho foi desenvolver uma ferramenta computacional Web para roteamento de veículos, denominada SGRV  Sistema de Gestão de Roteamento de Veículos, que emprega recursos do Google Maps e visa suprir as necessidades de microempresas. Para tanto, inicialmente realizou-se uma pesquisa bibliográfica acerca de métodos de solução para o PRV, a fim de subsidiar a escolha de uma estratégia baseada em Algoritmos Genéticos empregada no SGRV. Em seguida, foi feito um novo levantamento bibliográfico com intuito de encontrar softwares de uso livre para a solução do PRV, os quais foram objeto de uma análise que visou balizar a implementação do SGRV. Para o desenvolvimento dessa pesquisa foi empregada a metodologia Design Science Research. Uma avaliação qualitativa do SGRV foi realizada por quatro microempresas de diferentes ramos, as quais utilizaram a ferramenta por um determinado período e, ao final, responderam seis questões abertas a partir de entrevistas semiestruturadas. As experiências dessas microempresas foram transcritas e os dados obtidos revelam a efetividade do SGRV no gerenciamento de suas tarefas relacionadas aos pedidos e entregas de produtos.

A detailed family health history is currently the most potentially useful tool for diagnosis and risk assessment in clinical genetics. We developed and evaluated the usability and analytic validity of a patient-driven web-based family health history collection and analysis tool. Health Heritage © guides users through the collection of their family health history by relative, generates a pedigree, completes risk assessment, stratification, and recommendations for 89 conditions. We compared the performance of Health Heritage to that of Usual Care using a nonrandomized cohort trial of 109 volunteers. We contrasted the completeness and sensitivity of family health history collection and risk assessments derived from Health Heritage and Usual Care to those obtained by genetic counselors and genetic assessment teams. Nearly half (42%) of the Health Heritage participants reported discovery of health risks; 63% found the information easy to understand and 56% indicated it would change their health behavior. Health Heritage consistently outperformed Usual Care in the completeness and accuracy of family health history collection, identifying 60% of the elevated risk conditions specified by the genetic team versus 24% identified by Usual Care. Health Heritage also had greater sensitivity than Usual Care when comparing the identification of risks. These results suggest a strong role for automated family health history collection and risk assessment and underscore the potential of these data to serve as the foundation for comprehensive, cost-effective personalized genomic medicine.

<div>With the rapid development in high-throughput technologies and the next generation sequencing (NGS) during the past decades, the bottleneck for advances in computational biology and bioinformatics research has shifted from data collection to data analysis. As one of the central goals in precision health, understanding and interpreting high-dimensional biomedical data is of major interest in computational biology and bioinformatics domains. Since significant effort has been committed to harnessing biomedical data for multiple analyses, this thesis is aiming for developing new machine learning approaches to help discover and interpret the complex mechanisms and interactions behind the high dimensional features in biomedical data. Moreover, this thesis also studies the prediction of post-treatment response given histopathologic images with machine learning.</div><div><br></div><div>Capturing the important features behind the biomedical data can be achieved in many ways such as network and correlation analyses, dimensionality reduction, image processing, etc. In this thesis, we accomplish the computation through co-expression analysis, survival analysis, and matrix decomposition in supervised and unsupervised learning manners. We use co-expression analysis as upfront feature engineering, implement survival regression in deep learning to predict patient survival and discover associated factors. By integrating Cox proportional hazards regression into non-negative matrix factorization algorithm, the latent clusters of human genes are uncovered. Using machine learning and automatic feature extraction workflow, we extract thirty-six image features from histopathologic images, and use them to predict post-treatment response. In addition, a web portal written by R language is built in order to bring convenience to future biomedical studies and analyses.</div><div><br></div><div>In conclusion, driven by machine learning algorithms, this thesis focuses on the integrative analysis given multimodal biomedical data, especially the supervised cancer patient survival prognosis, the recognition of latent gene clusters, and the application of predicting post-treatment response from histopathologic images. The proposed computational algorithms present its superiority comparing to other state-of-the-art models, provide new insights toward the biomedical and cancer studies in the future.</div>

In the last decade, cyber-threats have evolved dramatically, forcing organizations yearafter year to use increasingly sophisticated security measures, security software amongothers. This has led to a huge increase in the number of security tools available in theindustry. The result of the increase is that that companies often do not know in whichsoftware to invest in order to meet their security needs. The purpose of this thesis isto address this problem by developing a solution that helps companies to choose theright security software based on their security needs and that allows to do the selectionprocess in a systematic and reliable way.The solution proposed in the thesis builds on interviews with experts in information security,data collection from the literature and Internet and on a case study. The solutionconsists of rstly an investigate method with which it is possible to categorize any securitytool according to the list of cyber-security controls proposed by CIS Critical SecurityControls (CSC), which were chosen after a comparative study with other publicly availablecontrols because they are actionable, relevant and updated frequently. Secondly,the solution proposes a user-friendly web tool that has been developed to allow the usersto visualize the collected information for comparison. The visualization tool will helpthe users to select the security tools in which the company could be interested to investin. The visualization is done in a simple way and the CSCs that would be covered areshown together with the gaps and the overlaps of the selected tools. In order to verifythe viability of the solution that was developed with real data, the project includes acase study with a representative set of security tools. The case study facilitates thecomprehension of the process undertaken and shows how this method could be appliedin a real case scenario. / Under det senaste decenniet har cyberhot utvecklats dramatiskt. Hotet tvingar organisationeratt år efter år använda allt mer sofistikerade säkerhetsåtgärder, bland annatsäkerhetsmjukvara. Detta har lett till en enorm ökning av antalet av säkerhetsverktygsom finns i branschen. Resultatet av ökningen är att företag ofta inte vet i vilken programvarade borde investera i för att möta sina säkerhetsbehov. Syftet med dennarapporten är att ta itu med detta problem genom att utveckla en lösning som hjälperföretag att välja rätt säkerhetsprogramvara baserat på deras säkerhetsbehov och somgör urvalsprocessen på ett systematiskt och tillförlitligt sätt.Den lösning som föreslås i rapporten bygger på intervjuer med experter inom informationssäkerhet, datainsamling från litteraturen och Internet och på en fallstudie. Lösningenbestår först av en utredningsmetod med vilken det är möjligt att kategorisera vilketsäkerhetsverktyg som helst enligt listan över cybersäkerhetskontroller som publiceras avCIS Critical Security Controls (CSC). CSC valdes efter en jämförande studie som inkluderadeandra allmänt tillgängliga förteckningar över kontrollerna, eftersom CSC kontrollerär genomförbara, relevanta och uppdateras ofta. För det andra föreslår lösningen ettanvändarvänligt webbverktyg som har utvecklats för att göra det möjligt för användareatt visualisera den insamlade informationen för jämförelse. Visualiseringsverktyget kommeratt hjälpa användarna välja säkerhetsverktyg som företaget kan vara intresseradeav att investera i. Visualiseringen sker på ett enkelt sätt och CSCs som omfattas visastillsammans med de luckor och överlappningar som finns i den valda programvaran.För att bekräfta genomförbarhet för den lösning som utvecklats med verkliga data,omfattar projektet en fallstudie med ett representativt urval av säkerhetsverktyg. Fallstudienunderlättar förståelsen för klassificeringen och urvalsprocessen genom att visahur denna metod skulle kunna tillämpas i ett verkligt fall.