Symbolic Knowledge Extraction from Trained Neural Networks Governed by Lukasiewicz Logics

This work describes a methodology to extract symbolic rules from trained neural networks. In our approach, patterns on the network are codified using formulas on a Lukasiewicz logic. For this we take advantage of the fact that every connective in this multi-valued logic can be evaluated by a neuron in an artificial network having, by activation function the identity truncated to zero and one. This fact simplifies symbolic rule extraction and allows the easy injection of formulas into a network architecture. We trained this type of neural network using a back-propagation algorithm based on Levenderg-Marquardt algorithm, where in each learning iteration, we restricted the knowledge dissemination in the network structure. This makes the descriptive power of produced neural networks similar to the descriptive power of Lukasiewicz logic language, minimizing the information loss on the translation between connectionist and symbolic structures. To avoid redundance on the generated network, the method simplifies them in a pruning phase, using the "Optimal Brain Surgeon" algorithm. We tested this method on the task of finding the formula used on the generation of a given truth table. For real data tests, we selected the Mushrooms data set, available on the UCI Machine Learning Repository.

Training control centers' operators in incident diagnosis and power restoration using intelligent tutoring systems

The activity of Control Center operators is important to guarantee the effective performance of Power Systems. Operators’ actions are crucial to deal with incidents, especially severe faults like blackouts. In this paper, we present an Intelligent Tutoring approach for training Portuguese Control Center operators in tasks like incident analysis and diagnosis, and service restoration of Power Systems. Intelligent Tutoring System (ITS) approach is used in the training of the operators, having into account context awareness and the unobtrusive integration in the working environment. Several Artificial Intelligence techniques were criteriously used and combined together to obtain an effective Intelligent Tutoring environment, namely Multiagent Systems, Neural Networks, Constraint-based Modeling, Intelligent Planning, Knowledge Representation, Expert Systems, User Modeling, and Intelligent User Interfaces.

Ontologies for reusing data cleaning knowledge

The emergence of new business models, namely, the establishment of partnerships between organizations, the chance that companies have of adding existing data on the web, especially in the semantic web, to their information, led to the emphasis on some problems existing in databases, particularly related to data quality. Poor data can result in loss of competitiveness of the organizations holding these data, and may even lead to their disappearance, since many of their decision-making processes are based on these data. For this reason, data cleaning is essential. Current approaches to solve these problems are closely linked to database schemas and specific domains. In order that data cleaning can be used in different repositories, it is necessary for computer systems to understand these data, i.e., an associated semantic is needed. The solution presented in this paper includes the use of ontologies: (i) for the specification of data cleaning operations and, (ii) as a way of solving the semantic heterogeneity problems of data stored in different sources. With data cleaning operations defined at a conceptual level and existing mappings between domain ontologies and an ontology that results from a database, they may be instantiated and proposed to the expert/specialist to be executed over that database, thus enabling their interoperability.

An architecture for ambient intelligent environments

This paper presents the proposal of an architecture for developing systems that interact with Ambient Intelligence (AmI) environments. This architecture has been proposed as a consequence of a methodology for the inclusion of Artificial Intelligence in AmI environments (ISyRAmI - Intelligent Systems Research for Ambient Intelligence). The ISyRAmI architecture considers several modules. The first is related with the acquisition of data, information and even knowledge. This data/information knowledge deals with our AmI environment and can be acquired in different ways (from raw sensors, from the web, from experts). The second module is related with the storage, conversion, and handling of the data/information knowledge. It is understood that incorrectness, incompleteness, and uncertainty are present in the data/information/knowledge. The third module is related with the intelligent operation on the data/information/knowledge of our AmI environment. Here we include knowledge discovery systems, expert systems, planning, multi-agent systems, simulation, optimization, etc. The last module is related with the actuation in the AmI environment, by means of automation, robots, intelligent agents and users.

Os projetos educativos enquanto manifestações da identidade das instituições e as suas implicações no estabelecimento de parcerias

Dissertação de Mestrado, Educação Pré-Escolar e Ensino do 1º Ciclo do Ensino Básico, 28 de Janeiro de 2013, Universidade dos Açores (Relatório de Estágio).

O Comportamento de consumo de Vinho do Porto dos portugueses – recomendações para o aumento da procura interna

Dissertação de Mestrado apresentada ao Instituto de Contabilidade e Administração do Porto para a obtenção do grau de Mestre em Marketing Digital, sob orientação de Doutor António Correia de Barros

Gestão fiscal e benefícios fiscais

Dissertação de Mestrado apresentado ao Instituto de Contabilidade e Administração do Porto para a obtenção do grau de Mestre em Auditoria, sob orientação de Doutor José Campos Amorim

A relação terapeuta-cliente: na perspectiva do cliente

O presente estudo resulta de uma crescente preocupação e percepção da importância da relação terapeuta-cliente e procura compreender a perspectiva do cliente e a forma como este sente e vive essa relação ao longo de todo o seu processo terapêutico. O estudo realizado, de natureza qualitativa e de carácter exploratório, visa sobretudo analisar a influência da relação terapeuta-cliente na perspectiva do cliente, identificar e descrever os fatores que a influenciam e perceber a mais-valia desta relação ao longo do processo terapêutico. Para tal, foram entrevistadas dezanove pessoas que estão a receber tratamento de terapia ocupacional. Para a recolha dos dados foi aplicada a entrevista semi-estruturada por se pretender um contributo mais envolvente, particularizado e subjetivo dos clientes. Através da realização do estudo e de acordo com a perspetiva dos clientes entrevistados, concluiu-se que a capacidade do terapeuta em construir e estabelecer um vínculo com o cliente ditará grande parte do sucesso ou insucesso do processo terapêutico. Porém, tão importante como formar e estabelecer um vínculo terapeuta-cliente é preciso ter a preocupação contínua de o manter sempre vivo e fortalecido. Para que tal aconteça, o terapeuta não pode menosprezar a experiência de vida e expectativas do cliente e deve assumir um papel de permanente preocupação e atenção a todas as transformações quotidianas pois estas influenciam o envolvimento do cliente nas suas atividades/ocupações e afetam a sua saúde e desempenho. O novo milénio requer dos profissionais de saúde em geral e dos terapeutas ocupacionais em particular, novas habilidades e competências. É fundamental adquirir a consciência de que os clientes são o centro de todo o processo terapêutico. É necessário ter uma visão holística e não fragmentada do cliente. É importante interagir com os clientes e permitir uma troca de conhecimento, entre o saber do cliente e o saber do terapeuta. Essa troca gera convergências, fortalece laços e o processo terapêutico avança. Nesse relacionamento, ambos aprendem, progridem e crescem.

Desenvolvimento lexical: perspectivas e práticas de professores no 1.º ciclo do ensino básico

Dissertação apresentada à Escola Superior de Educação de Lisboa para a obtenção de grau de Mestre em Didática da Língua Portuguesa no 1.º e 2.º Ciclos do Ensino Básico

A dimensão política da segurança para o ciberespaço na União Europeia : a agenda digital, a estratégia de cibersegurança e a cooperação UE-OTAN

Mestrado em Relações Internacionais.

Knowledge, attitude and practice of mammography among women users of public health services

OBJECTIVE: To evaluate knowledge, attitude and practice related to mammography among women users of local health services, identifying barriers to its performance. METHODS: A total of 663 women were interviewed at 13 local health centers in a city of Southeastern Brazil, in 2001. Interviewees were randomly selected at each center and they were representative from different socioeconomic conditions. The number of interviewees at each center was proportional to monthly mean appointments. For data analysis, answers were described as knowledge, attitude, practice and their respective adequacies and then they were correlated with control variables through the chi-square test. RESULTS: Only 7.4% of the interviewees had adequate knowledge on mammography, while 97.1% of women had an adequate attitude. The same was seen for the practice of mammography that was adequate in 35.7% of the cases. The main barrier to mammography was lack of referral by physicians working at the health center (81.8%). There was an association between adequacy of attitude and five years or more of education and being married. There was also an association between adequacy of mammography practice and being employed and family income up to four minimum wages. CONCLUSIONS: Women users of local health services had no adequate knowledge and practice related to mammography despite having an adequate attitude about this exam.

O impacto do ensino de estratégias morfológicas no desenvolvimento da escrita: um estudo de intervenção

Dissertação de mestrado em ciências da educação especialidade educação especial

Management of agricultural waste on a sample of farms on the island of São Miguel (Açores)

Dissertação de Mestrado em Ambiente, Saúde e Segurança.

HIV testing among pregnant women in Brazil: rates and predictors

OBJECTIVE: To assess rates of offering and uptake of HIV testing and their predictors among women who attended prenatal care. METHODS: A population-based cross-sectional study was conducted among postpartum women (N=2,234) who attended at least one prenatal care visit in 12 cities. Independent and probabilistic samples were selected in the cities studied. Sociodemographic data, information about prenatal care and access to HIV prevention interventions during the current pregnancy were collected. Bivariate and multivariate analyses were carried out to assess independent effects of the covariates on offering and uptake of HIV testing. Data collection took place between November 1999 and April 2000. RESULTS: Overall, 77.5% of the women reported undergoing HIV testing during the current pregnancy. Offering of HIV testing was positively associated with: previous knowledge about prevention of mother-to-child transmission of HIV; higher number of prenatal care visits; higher level of education and being white. HIV testing acceptance rate was 92.5%. CONCLUSIONS: The study results indicate that dissemination of information about prevention of mother-to-child transmission among women may contribute to increasing HIV testing coverage during pregnancy. Non-white women with lower level of education should be prioritized. Strategies to increase attendance of vulnerable women to prenatal care and to raise awareness among health care workers are of utmost importance.

Potencialidades e utilização do espaço recreio: um estudo desenvolvido em escolas do 1.º ciclo do ensino básico

Dissertação apresentada para a obtenção do grau de Mestre em Educação - Área de Especialização em Didática das Ciências