Dissertação (mestrado)—Universidade de Brasília, Faculdade Gama, Programa de Pós-Graduação em Engenharia Biomédica, 2017. / Submitted by Fernanda Percia França (fernandafranca@bce.unb.br) on 2017-05-11T17:09:26Z No. of bitstreams: 1 2017_HendersonMatsuuraSanches.pdf: 3859884 bytes, checksum: 321d23329465fde2ebd5422eaa30e5a4 (MD5) / Approved for entry into archive by Raquel Viana (raquelviana@bce.unb.br) on 2017-05-16T21:37:03Z (GMT) No. of bitstreams: 1 2017_HendersonMatsuuraSanches.pdf: 3859884 bytes, checksum: 321d23329465fde2ebd5422eaa30e5a4 (MD5) / Made available in DSpace on 2017-05-16T21:37:03Z (GMT). No. of bitstreams: 1 2017_HendersonMatsuuraSanches.pdf: 3859884 bytes, checksum: 321d23329465fde2ebd5422eaa30e5a4 (MD5) Previous issue date: 2017-05-16 / O objetivo desse trabalho foi a construção de um modelo ontológico da Neoplasia Mamária (NM) denominado ONTO-MAMA-NM. Esse modelo é uma ferramenta importante para auxiliar especialistas e estudantes da área da saúde no tratamento do câncer de mama. O modelo ontológico foi criado na linguagem Web Ontology Language (OWL), cuja principal vantagem é a facilidade para expressar significados e semântica e aplicabilidade no processo de informações de forma automatizada. Por se tratar de um modelo aplicado à área médica, o ONTO-MANA-NM procura manter a compatibilidade com os padrões Digital Imaging and Communications in Medicine (DICOM) e Health Level Seven International (HL7), de modo a preservar a interoperabilidade das informações dos pacientes em ambientes hospitalares. Como resultado, obteve-se um detalhamento da ontologia desenvolvida e implementada no software Protégé 5.1 com o apoio da metodologia denominada de Methontology. Foi descrito todo o processo de desenvolvimento, desde a coleta de dados até a validação final do modelo junto aos especialistas. Sendo assim, foi avaliado em duas etapas, isto é, primeiramente pelos especialistas: fisioterapeutas, médicos, residentes e alunos da fisioterapia e medicina do HUB. Ao final do processo da validação do ONTO-MAMANM, informaram que desconheciam a ontologia e não tinham visto nada semelhante referente ao tratamento da NM, obtendo assim o primeiro modelo ontológico do tratamento da NM. / The aim of this work was the development of a Mammary Neoplasia (NM) ontological model called ONTOMAMA- NM. This model is a relevant tool to assist experts and students of the health area in the treatment of breast cancer. The ontological model was implemented in the Web Ontology Language (OWL) language, whose main advantage is the facility to express meanings, semantics and applicability in the information process in an automated way. As a model applied to the medical field, ONTO-MANA-NM seeks to maintain compatibility with the Digital Imaging and Communications in Medicine (DICOM) and Health Level Seven International (HL7) standards, in order to preserve the interoperability of patients information in hospital environments. As a result, it was developed a detailed ontology and implemented in Protégé 5.1 software with the support of the methodology called as Methontology. The final development process was described since the data collection until the final validation of the model with the experts. Thus, it was evaluated in two stages, that is, firstly by the specialists: physiotherapists, physicians, residents and students of physiotherapy and HUB medicine. At the end of the ONTO-MAMA-NM validation process, they reported that they did not know about the ontology and had not seen anything similar regarding NM treatment, thus obtaining the first ontological model of NM treatment.

Dissertação (mestrado)—Universidade de Brasília, Instituto de Ciências Exatas, Departamento de Ciência da Computação, Programa de Pós-Graducação em Informática, 2016. / Submitted by Fernanda Percia França (fernandafranca@bce.unb.br) on 2016-08-15T19:28:04Z No. of bitstreams: 1 2016_LaécioLimadosSantos.pdf: 8101789 bytes, checksum: cfa3494dd6c9cb6bcef19bad2ae8150d (MD5) / Approved for entry into archive by Raquel Viana(raquelviana@bce.unb.br) on 2016-10-11T21:17:42Z (GMT) No. of bitstreams: 1 2016_LaécioLimadosSantos.pdf: 8101789 bytes, checksum: cfa3494dd6c9cb6bcef19bad2ae8150d (MD5) / Made available in DSpace on 2016-10-11T21:17:42Z (GMT). No. of bitstreams: 1 2016_LaécioLimadosSantos.pdf: 8101789 bytes, checksum: cfa3494dd6c9cb6bcef19bad2ae8150d (MD5) / A Web Semântica (WS) adiciona informações semânticas a Web tradicional, permitindo que os computadores entendam conteúdos antes acessíveis apenas aos humanos. A Ontology Web Language (OWL), linguagem padrão para criação de ontologias na WS, se baseia em lógica descritiva para permitir uma modelagem formal de um domínio de conhecimento. A OWL, no entanto, não possui suporte para tratamento de incerteza, presente em diversas situações, o que motivou o estudo de várias alternativas para tratar este problema. O Probabilistic OWL (PR-OWL) adiciona suporte à incerteza ao OWL utilizando Multi-Entity Bayesian Networks (MEBN), uma linguagem probabilística de primeira ordem. A inferência no MEBN ocorre através da geração de uma rede bayesiana específica de situação (SSBN). O PR-OWL 2 estende a linguagem original oferecendo uma maior integração com o OWL e permitindo a construção de ontologias que mesclam conhecimento determinístico e probabilístico. PR-OWL não permite lidar com domínios que contenham bases assertivas muito grandes. Isto se deve a alta complexidade computacional da lógica descritiva na qual a OWL é baseada e ao fato de que as máquinas de inferência utilizadas nas implementações das versões do PR-OWL requerem que a base assertiva esteja carregada em memória. O presente trabalho propõe o PR-OWL 2 RL, uma versão escalável do PR-OWL baseada no profile OWL 2 RL e em triplestores. O OWL 2 RL permite raciocínio em tempo polinomial para as principais tarefas de inferência. Triplestores permitem armazenar triplas RDF (Resource Description Framework) em bancos de dados otimizados para trabalhar com grafos. Para permitir a geração de SSBN para bases contendo muitas evidências, este trabalho propõe um novo algoritmo, escalável ao instanciar nós de evidência apenas caso eles influenciem o nó objetivo. O plug-in PR-OWL 2 RL para o framework UnBBayes foi desenvolvido para permitir uma avaliação experimental dos algoritmos propostos. O estudo de caso abordado foi o de fraudes em licitações públicas. _______________________________________________________________________________________________ ABSTRACT / Semantic Web (SW) adds semantic information to the traditional Web, allowing computers to understand content before accessible only by human beings. The Web Ontology Language (OWL), main language for building ontologies in SW, allows a formal modeling of a knowledge domain based on description logics. OWL, however, does not support uncertainty. This restriction motivated the creation of several extensions of this language. Probabilistic OWL (PROWL) improves OWL with the ability to treat uncertainty using Multi-Entity Bayesian Networks (MEBN). MEBN is a first-order probabilistic logic. Its inference consists of generating a Situation Specific Bayesian Network (SSBN). PR-OWL 2 extends the PR-OWL offering a better integration with OWL and its underlying logic, allowing the creation of ontologies with deterministic and probabilistic parts. PR-OWL, however, does not deal with very large assertive bases. This is due to the high computational complexity of the description logic of OWL. Another fact is that reasoners used in PR-OWL implementation require that the data be fully load into memory at the time of inference. To address this issue, this work proposes PR-OWL 2 RL, a scalable version of PR-OWL based on OWL 2 RL profile and on triplestores. OWL 2 RL allows reasoning in polynomial time for the main reasoning tasks. Triplestores can store RDF (Resource Description Framework) triples in databases optimized to work with graphs. To allow the generation of SSBNs for databases with large evidence base, this work proposes a new algorithm that is scalable because it instantiates an evidence node only if it influence a target node. A plug-in for the UnBBayes framework was developed to allow an empirical evaluation of the new algorithms proposed. A case study over frauds into procurements was carried on.

M. Tech. (Department of Information & Communication Technology, Faculty of Applied and Computer Sciences), Vaal University of Technology. / Many studies have evaluated and compared the existing open-sources Semantic Web platforms for ontologies development. However, none of these studies have included the dot NET-based semantic web platforms in the empirical investigations. This study conducted a comparative analysis of open-source and dot NET-based semantic web platforms for ontologies development. Two popular dot NET-based semantic web platforms, namely, SemWeb.NET and dotNetRDF were analyzed and compared against open-source environments including Jena Application Programming Interface (API), Protégé and RDF4J also known as Sesame Software Development Kit (SDK). Various metrics such as storage mode, query support, consistency checking, interoperability with other tools, and many more were used to compare two categories of platforms. Five ontologies of different sizes are used in the experiments. The experimental results showed that the open-source platforms provide more facilities for creating, storing and processing ontologies compared to the dot NET-based tools. Furthermore, the experiments revealed that Protégé and RDF4J open-source and dotNetRDF platforms provide both graphical user interface (GUI) and command line interface for ontologies processing, whereas, Jena open-source and SemWeb.NET are command line platforms. Moreover, the results showed that the open-source platforms are capable of processing multiple ontologies’ files formats including Resource Description Framework (RDF) and Ontology Web Language (OWL) formats, whereas, the dot NET-based tools only process RDF ontologies. Finally, the experiment results indicate that the dot NET-based platforms have limited memory size as they failed to load and query large ontologies compared to open-source environments.