An Extension of Estimation of Domain of Attraction for Fractional Order Linear System Subject to Saturation Control

This paper employs the Lyapunov direct method for the stability analysis of fractional order linear systems subject to input saturation. A new stability condition based on saturation function is adopted for estimating the domain of attraction via ellipsoid approach. To further improve this estimation, the auxiliary feedback is also supported by the concept of stability region. The advantages of the proposed method are twofold: (1) it is straightforward to handle the problem both in analysis and design because of using Lyapunov method, (2) the estimation leads to less conservative results. A numerical example illustrates the feasibility of the proposed method.

Towards the Certification of Multicore Platforms in the Avionics Domain

Poster presented in Work in Progress Session, The 28th GI/ITG International Conference on Architecture of Computing Systems (ARCS 2015). 24 to 27, Mar, 2015. Porto, Portugal.

Towards the Certification of Multicore Platforms in the Avionics Domain

Presented at Work in Progress Session, The 28th GI/ITG International Conference on Architecture of Computing Systems (ARCS 2015). 24 to 27, Mar, 2015. Porto, Portugal.

Interactional strategies between terminologist and domain specialist. Putting order into our universe: the concept of Blended Learning

A presente comunicação visa discutir as mais-valias de um desenho metodológico sustentado numa abordagem conceptual da Terminologia aplicado ao exercício de harmonização da definição do cenário educativo mais promissor do Ensino Superior actual: o blended learning. Sendo a Terminologia uma disciplina que se ocupa da representação, da descrição e da definição do conhecimento especializado através da língua a essência deste domínio do saber responde a uma necessidade fundamental da sociedade actual: putting order into our universe, nas palavras de Nuopponen (2011). No contexto descrito, os conceitos, enquanto elementos da estrutura do conhecimento (Sager, 1990) constituem um objecto de investigação de complexidade não despicienda, pois apesar do postulado de que a língua constitui uma ferramenta fundamental para descrever e organizar o conhecimento, o princípio isomórfico não pode ser tomado como adquirido. A abordagem conceptual em Terminologia propõe uma visão precisa do papel da língua no trabalho terminológico, sendo premissa basilar que não existe uma correspondência unívoca entre os elementos atomísticos do conhecimento e os elementos da expressão linguística. É pela razões enunciadas que as opções metodológicas circunscritas à análise do texto de especialidade serão consideradas imprecisas. Nesta reflexão perspectiva-se que o conceito-chave de uma abordagem conceptual do trabalho terminológico implica a combinação de um processo de elicitação do conhecimento tácito através de uma negociação discursiva orientada para o conceito e a análise de corpora textuais. Defende-se consequentemente que as estratégias de interacção entre terminólogo e especialista de domínio merecem atenção detalhada pelo facto de se reflectirem com expressividade na qualidade dos resultados obtidos. Na sequência do exposto, o modelo metodológico que propomos sustenta-se em três etapas que privilegiam um refinamento dessa interacção permitindo ao terminólogo afirmar-se como sujeito conceptualizador, decisor e interventor: (1) etapa exploratória do domínio-objecto de estudo; (2) etapa de análise onamasiológica de evidência textual e discursiva; (3) etapa de modelização e de validação de resultados. Defender-se-á a produtividade de uma sequência cíclica entre a análise textual e discursiva para fins onomasiológicos, a interacção colaborativa e a introspecção.

Multipath inter-domain policy routing

Dissertação submetida para a obtenção do grau de Doutor em Engenharia Electrotécnica e de Computadores

Semantic web approach for dealing with administrative boundary revisions: a case study of Dhaka City

Dissertation submitted in partial fulfillment of the requirements for the Degree of Master of Science in Geospatial Technologies.

Child programming: an adequate domain specific language for programming specific robots

Dissertação para obtenção do Grau de Mestre em Engenharia Informática

Barreiras e facilitadores à implementação de um modelo de gestão de cuidados em Saúde Mental

RESUMO: Enquadramento teórico - Os estudos epidemiológicos demonstram que apesar de todo o progresso científico, muitas pessoas continuam sem acesso aos Serviços de Saúde Mental (SSM) e que, em muitos casos, os cuidados não têm a qualidade suficiente. A experiência de vários países mostra que os processos de implementação de modelos de intervenção terapêutica, como é o da Gestão de Cuidados, são lentos e complexos, não dependendo somente do grau de efectividade ou da complexidade das práticas a implementar. O Modelo de Gestão de Cuidados (MGC), é definido como uma prática baseada na evidência, utilizada para ajudar os doentes nos seus processos de recuperação. As estratégias para implementar práticas baseadas na evidência são críticas para a melhoria dos serviços. Existem, apesar de toda a evidência, muitas barreiras à implementação. Ao constatarmos que as práticas validadas pela ciência estão longe de estar claramente disseminadas nos serviços de saúde mental, fundamentamos a necessidade de utilizar metodologia de implementação que, além da efectividade das práticas, permita uma efectividade da implementação. Para responder às necessidades de formação e no âmbito da implementação do Plano Nacional de Saúde Mental, foram formados, em Portugal, 170 profissionais de saúde mental provenientes de serviços públicos e do sector social, de todas as regiões de Portugal Continental. Considerando que estes profissionais adquiriram competências específicas no MGC, através de um programa de formação nacional idêntico para todos os serviços de saúde mental, investigámos o grau de implementação deste modelo, bem como os facilitadores e as barreiras à sua correcta implementação. Existem vários estudos internacionais sobre as barreiras e os facilitadores à implementação de práticas baseadas na evidência, embora a maior parte desses estudos seja baseado em entrevistas semi-estruturadas a profissionais. Por outro lado, não existem, em Portugal, estudos sobre as barreiras e os facilitadores à implementação de práticas de saúde mental. Objectivos 1. Estimar o grau da implementação do MGC nos serviços de saúde mental portugueses 2. Caracterizar as regiões onde a implementação do MGC tenha ocorrido em maior grau. 3. Identificar os factores facilitadores e as barreiras à implementação do MGC, entre as regiõesde saúde do país. 4. Explorar as relações entre a fidelidade da implementação, as barreiras e os facilitadores da implementação, a cultura organizacional e as características dos serviços de saúde mental. Metodologia Estudo observacional, transversal e descritivo, com características exploratórias. População: profissionais dos serviços de saúde mental públicos e do sector social que frequentaram o Programa Nacional de Formação em Saúde Mental Comunitária no curso “Cuidados Integrados e Recuperação”, da Coordenação Nacional para a Saúde Mental / Ministério da Saúde, entre Outubro de 2008 e Dezembro de 2009, (n=71). Avaliação Fidelidade de implementação do Modelo de Gestão de Cuidados - IMR-S (Illness Management and Recovery Scale); Qualidade das guidelines utilizadas na implementação do Modelo de Gestão de Cuidados - AGREE II-PT (Appraisal of Guidelines, for Research and Evaluation); Avaliação das Barreiras e Facilitadores à implementação do MGC - BaFAI (Barriers and Facilitators Assessment Instrument); Avaliação da Cultura Organizacional dos serviços de saúde mental - CVF-I (Competing Values Framework Instrument). Análise Estatística Para a descrição dos dados foram aplicados métodos de estatística descritiva. Para a comparação de subgrupos foram utilizados os testes de Mann Whitney e Kruskall-Wallis. Para a investigação de associações foram utilizados os métodos de correlação de Spearman e a Regressão Múltipla. O tratamento e análise dos dados foram realizados utilizando o programa estatístico IBM SPSS Statistics® para Mac/Apple® nas versões 19 e 20. Resultados Serviços: A articulação com os cuidados de saúde primários existe na maioria dos serviços (56.34%) e 77.46% dos serviços têm autonomia para definir os cuidados a prestar. A maioria dos serviços (63.38%) realiza duas ou mais reuniões clínicas por mês e a quase totalidade (95.77%) recebe estagiários e/ou internos. A área da investigação tem níveis considerados baixos, quando comparados com outros países da Europa, tanto para a globalidade das áreas de investigação (25.35%), como para as áreas psicossociais (22.54%). Considerando componentes fundamentais para a implementação de modelos de gestão de cuidados, os resultados nacionais indicam que 66.20% dos serviços fazem registos em processo clínico único. As percentagens de utilização de planos individuais de cuidados são globalmente baixas (46.48%). Por seu turno, a utilização de guidelines, nos serviços do país, tem uma percentagem média nacional de 57.75%. Profissionais: São, na sua maioria, do sexo feminino (69.01%), com idades entre os 25 e os 56 anos (média 38.9, ± 7.41). Pertencem, maioritariamente, aos grupos profissionais da enfermagem (23.94%) e da psicologia (49.30%). A formação dos profissionais é de nível superior em todos os grupos, com uma percentagem total de licenciados de 80.3%, tendo os restantes uma formação ao nível do mestrado. Apesar dos valores baixos (17%) de formação prévia em modelos de gestão de cuidados, 39% dos profissionais indicou utilizar algumas vertentes destes modelos na sua prática. Apesar de 97,18% dos profissionais ter participado em dois ou mais encontros científicos, num período de dois anos, apenas 38.03% apresentou alguma comunicação científica no mesmo intervalo. Guideline: Os resultados da avaliação da guideline do MGC indicaram percentagens mais altas, quanto à qualidade do seu desenvolvimento, nos Domínios 1 (Objectivo e finalidade, com 72.2%) e 4 (Clareza de Apresentação, 77.7%). O Domínio 5 (Aplicabilidade) foi pontuado no limite inferior do desenvolvimento com qualidade suficiente (54.1%), ao passo que a guideline obteve uma pontuação negativa nos Domínios 2 (Envolvimento das partes interessadas, com 41.6%) e 3 (Rigor do Desenvolvimento, com 28.1%). Adicionalmente não foi possível às avaliadoras cotar o Domínio 6 (Independência editorial), por ausência de referências neste contexto. A guideline teve uma avaliação global positiva (66%), com recomendação de aceitação com modificações. Cultura Organizacional: O perfil de liderança com maior frequência nos serviços de saúde mental portugueses foi o de Mentor (45.61%). As percentagens mais baixas pertenceram aos perfis Monitor e Inovador (3.51%). Na perspectiva da cultura organizacional dos serviços, apontuação mais alta foi a da Cultura das Relações Humanas (74.07%). A estratégia de liderança, com predomínio em todas as regiões, foi a estratégia de Flexibilidade (66.10%). Os resultados mostram que a única associação positivamente significativa com o grau da implementação do MGC é a do perfil Produtor, com um peso específico de 14.55% na prevalência dos perfis de liderança nos serviços de saúde mental portugueses. Barreiras: As barreiras à implementação da prática do MGC, identificadas pelos profissionais dos serviços de saúde mental, com percentagens mais altas nos totais do país, foram: o tempo (57.7%), o conhecimento sobre o modelo e a motivação (40.8%), a colaboração dos outros profissionais (33.7%), o número de contactos reduzidos com os doentes (35.2%), as insuficiências do ponto de vista dos espaços (70.4%) e dos instrumentos disponíveis (69%) para implementar o MGC. Existiu uma variação entre as regiões de saúde do país. Os resultados mostram que houve uma correlação negativa, de forma significativa, entre a implementação do MGC e as barreiras: da resistência à utilização de protocolos, do formato da prática, da necessidade de mais treino e da não cooperação dos profissionais. Foram encontradas diferenças estatisticamente significativas entre as barreiras à implementação e as características dos serviços, dos profissionais e da cultura organizacional. Implementação: A média nacional da fidelidade de implementação do MGC (41.48) teve valores aproximados aos de estudos similares. Na pontuação por regiões, a implementação com maior fidelidade ocorreu no Alentejo. Se considerarmos a implementação com fidelidade esta ocorreu em 57.75% dos serviços e uma boa implementação em 15.49%. Os métodos de regressão permitiram confirmar a capacidade preditiva das barreiras e da cultura organizacional quanto à fidelidade da implementação do MGC. Discussão: No universo das hipóteses inicialmente colocadas foi possível verificar a variação da implementação do MGC entre as regiões do país. O estudo permitiu, adicionalmente, concluir pela existência de denominadores comuns de maior sucesso da implementação do MGC. Foi ainda possível verificar uma relação significativa, existente entre o grau de implementação e as dimensões das barreiras, a cultura organizacional e os recursos dos SSM (aqui definidos pelas características dos serviços e dos profissionais). De uma forma mais conclusiva podemos afirmar que existem outros factores, que não estão relacionados com a avaliação restrita dos recursos financeiros ou humanos, associados à qualidade da implementação de práticas baseadas na evidência, como o MGC. Exemplo disso são os achados referentes à região de saúde do Alentejo, onde a distância dos grandes centros urbanos e as conhecidas dificuldades de acessibilidade, combinadas com os problemas conhecidos da falta de recursos, não impediram que fosse a região com os valores mais altos da fidelidade de implementação. Conclusões: Foram encontradas inúmeras barreiras à implementação do MGC. Existem barreiras diferentes entre regiões, que resultam das características dos serviços, dos profissionais e da cultura organizacional. Para existir implementação é necessária a consideração de metodologias próprias que vão para além dos tradicionais programas de formação. As práticas baseadas na evidência, amplamente defendidas, exigem implementações baseadas na evidência.-------------ABSTRACT: Introduction - Several epidemiological studies show that, despite all scientific progress, many people still continue to have no access to mental health services and in many situations the quality of care is poor. The experiences of several countries show that progress towards case management implementation is slow and complex, depending not only from the degree of effectiveness or the complexity of the practice. Case management is defined as an evidence-based practice used to help patients in the recovery process. Strategies to implement evidence-based practices are critical to services improvement. There are many barriers to their implementation, despite all available evidence. Realising that practices of proved scientific value are far from being clearly implemented, justifies the need to use implementation methodologies that, beyond practice effectiveness, allow implementation effectiveness. To answer training needs and in the framework of the National Mental Health Plan implementation, 170 mental health (MH) professionals from portuguese public and private sectors were trained. Considering that case management skills were acquired, as a result of this training programme, we decided to study the degree of implementation in the services.Barriers and facilitators to the implementation were studied as well. There are several studies related with barriers and facilitators to the implementation of evidence-based practices, but most of them use semi-structured interviews with professionals. Additionally, there are no studies in Portugal related with barriers and facilitators to the implementation of mental health practices. Objectives1. Estimate the degree of case management implementation in Portuguese MH Services. 2.Describe regions where implementation occurred with higher fidelity degree. 3. Identify barriers and facilitators to case management implementation across country regions. 4. Explore the relationships between implementation, barriers and facilitators, organisational culture and services characteristics. Methodology - Cross sectional, descriptive study. Assessments - Implementation fidelity - IMR-S (Illness Management and Recovery Scale); Guideline quality - AGREE II-PT (Appraisal of Guidelines, for Research and Evaluation); Barriers and facilitators assessment - BaFAI (Barriers and Facilitators Assessment Instrument); Organisational culture assessment - CVF-I (Competing Values Framework Instrument). Statistical analysis - Descriptives and cross-tabs. Subgroups comparison: Mann-Witney and Kruskall-Wallis. Associations between variables were calculated using Spearman correlation's and Multiple Regression. Results - Services: Liaison with primary care is done in most services (56.34%) and 77.46% have autonomy to determine care. Most services have regular clinical meetings and almost all give internship training (95.77%). Research activity is low compared with other European countries, for both general and psychosocial research. Considering key components for the case management implementation, 66.20% of all services use single clinical records. The use of individual care plans is globally low (46.48%) and there is a use of guidelines in 57.75% of services. Human Resources: most are women (69.01%), with age ranging from 25-56 (average 39.9, SD 7.41). The majority are psychologists (49.30%) and nurses (23.94%). All have a university degree, 19.7% have a masters degree and 83% didn’t have any case management training before the above mentioned national training. Despite the low levels of preceding case management training, 39% have used model components in day-to-day practice and although 97.18% of the workforce have attended scientific meetings in the last 2 years, only 38.03% presented communications in the same period. Guideline: Results show that higher scores were obtained in Domain 1. Scope and Purpose (72.2.%),and Domain 4. Clarity of presentation (77.7%). Domain 5. pplicability scored near low boundary (54.1%) and negative scores were found in Domain 2. Stakeholder Involvement (41.6%) and Domain 3. Rigour of Development (28.1%). Global score was 66% and the guideline was recommended with modifications. Organisational Culture: The most frequent leadership profile was the Mentor profile (45.61%). Lower scores belonged to Innovator and Monitor profiles (3.51%). On the organisational culture overall, higher scores were found in the Human Relations culture (74.07%). The higher leadership strategy was the strategy of flexibility (66.10%). The results additionally showed that the only leadership profile associated with case management implementation was the Producer profile, representing 14.55% of all leadership profiles in the country.Barriers: The barriers identified by MH professionals, with high percentages, were: lack of time (57.7%), knowledge and motivation (40.8%), other colleagues cooperation (33.7%), low number of contacts with patients (35.2%), lack of facilities (70.4%) and lack of instruments (69%) to implement case management, varying across regions. Results show that there was a negative correlation between implementation and the following barriers: using protocols, practice format, need for more training and lack of cooperation from colleagues. Additionally, statistical differences were found between barriers to implementation and: services characteristics, workforce characteristics, organisational culture. Implementation: The national average results of case management implementation fidelity was (41.48), close to values found in similar studies. In the regional scores South Region Alentejo had the highest implementation score. If we look at minimum scores to assume implementation fidelity, these occurred in 57.75% of services and a good implementation occurred in 15.49% of these. Regression methods allowed to confirm that implementation score prediction was possible using the combination of barriers and organisational culture scores. Discussion - Considering the initial study hypotheses, it was possible to confirm the variation of case management implementation across country regions. Additionally, we could conclude that common denominators exist when successful implementation occurred. It was possible to observe a significant relationship between implementation degree and the dimensions of barriers, organisational culture and services resources (defined as professionals and services characteristics). In a more conclusive way, we can say that there are factors, other than financial and human resources, that are associated with evidence based practices implementation like case management. An example is the Alentejo region, were the distance from urban centres, and the known difficulties associated with accessibility, plus the lack of financial and human resources, have not impeded the regional higher score on implementation. Conclusions: Case management implementation had several barriers to implementation. There are different barriers across country regions, resulting from organisational culture, services and professionals characteristics. To reach implementation it is necessary to consider specific methodologies that go beyond traditional training programs and evident practices, widely promoted. Evidence-based practices require evidence-based implementations.

Semantic enrichment of knowledge sources supported by domain ontologies

This thesis introduces a novel conceptual framework to support the creation of knowledge representations based on enriched Semantic Vectors, using the classical vector space model approach extended with ontological support. One of the primary research challenges addressed here relates to the process of formalization and representation of document contents, where most existing approaches are limited and only take into account the explicit, word-based information in the document. This research explores how traditional knowledge representations can be enriched through incorporation of implicit information derived from the complex relationships (semantic associations) modelled by domain ontologies with the addition of information presented in documents. The relevant achievements pursued by this thesis are the following: (i) conceptualization of a model that enables the semantic enrichment of knowledge sources supported by domain experts; (ii) development of a method for extending the traditional vector space, using domain ontologies; (iii) development of a method to support ontology learning, based on the discovery of new ontological relations expressed in non-structured information sources; (iv) development of a process to evaluate the semantic enrichment; (v) implementation of a proof-of-concept, named SENSE (Semantic Enrichment kNowledge SourcEs), which enables to validate the ideas established under the scope of this thesis; (vi) publication of several scientific articles and the support to 4 master dissertations carried out by the department of Electrical and Computer Engineering from FCT/UNL. It is worth mentioning that the work developed under the semantic referential covered by this thesis has reused relevant achievements within the scope of research European projects, in order to address approaches which are considered scientifically sound and coherent and avoid “reinventing the wheel”.

A domain specific language for domotic systems

To cope with modernity, the interesting of having a fully automated house has been increasing over the years, as technology evolves and as our lives become more stressful and overloaded. An automation system provides a way to simplify some daily tasks, allowing us to have more spare time to perform activities where we are really needed. There are some systems in this domain that try to implement these characteristics, but this kind of technology is at its early stages of evolution being that it is still far away of empowering the user with the desired control over a habitation. The reason is that the mentioned systems miss some important features such as adaptability, extension and evolution. These systems, developed from a bottom-up approach, are often tailored for programmers and domain experts, discarding most of the times the end users that remain with unfinished interfaces or products that they have difficulty to control. Moreover, complex behaviors are avoided, since they are extremely difficult to implement mostly due to the necessity of handling priorities, conflicts and device calibration. Besides, these solutions are only reachable at very high costs, yet they still have the limitation of being difficult to configure by non-technical people once in runtime operation. As a result, it is necessary to create a tool that allows the execution of several automated actions, with an interface that is easy to use but at the same time supports all the main features of this domain. It is also desirable that this tool is independent of the hardware so it can be reused, thus a Model Driven Development approach (MDD) is the ideal option, as it is a method that follows those principles. Since the automation domain has some very specific concepts, the use of models should be combined with a Domain Specific Language (DSL). With these two methods, it is possible to create a solution that is adapted to the end users, but also to domain experts and programmers due to the several levels of abstraction that can be added to diminish the complexity of use. The aim of this thesis is to design a Domain Specific Language (DSL) that uses the Model Driven Development approach (MDD), with the purpose of supporting Home Automation (HA) concepts. In this implementation, the development of simple and complex scenarios should be supported and will be one of the most important concerns. This DSL should also support other significant features in this domain, such as the ability to schedule tasks, which is something that is limited in the current existing solutions.

Framework for the semantic alignment of enterprise’s domain knowledge

Nowadays, the consumption of goods and services on the Internet are increasing in a constant motion. Small and Medium Enterprises (SMEs) mostly from the traditional industry sectors are usually make business in weak and fragile market sectors, where customized products and services prevail. To survive and compete in the actual markets they have to readjust their business strategies by creating new manufacturing processes and establishing new business networks through new technological approaches. In order to compete with big enterprises, these partnerships aim the sharing of resources, knowledge and strategies to boost the sector’s business consolidation through the creation of dynamic manufacturing networks. To facilitate such demand, it is proposed the development of a centralized information system, which allows enterprises to select and create dynamic manufacturing networks that would have the capability to monitor all the manufacturing process, including the assembly, packaging and distribution phases. Even the networking partners that come from the same area have multi and heterogeneous representations of the same knowledge, denoting their own view of the domain. Thus, different conceptual, semantic, and consequently, diverse lexically knowledge representations may occur in the network, causing non-transparent sharing of information and interoperability inconsistencies. The creation of a framework supported by a tool that in a flexible way would enable the identification, classification and resolution of such semantic heterogeneities is required. This tool will support the network in the semantic mapping establishments, to facilitate the various enterprises information systems integration.

Production and diagnostic application of recombinant domain III of West Nile envelope protein in Brazil

INTRODUCTION: West Nile virus (WNV) is a flavivirus with a natural cycle involving mosquitoes and birds. Over the last 11 years, WNV has spread throughout the Americas with the imminent risk of its introduction in Brazil. METHODS: Envelope protein domain III of WNV (rDIII) was bacterially expressed and purified. An enzyme-linked immunosorbent assay with WNV rDIII antigen was standardized against mouse immune fluids (MIAFs) of different flavivirus. RESULTS: WNV rDIII reacted strongly with St. Louis encephalitis virus (SLEV) MIAF but not with other flaviviruses. CONCLUSIONS: This antigen may be a potentially useful tool for serologic diagnosis and may contribute in future epidemiological surveillance of WNV infections in Brazil.

Leishmaniasis transmission: distribution and coarse-resolution ecology of two vectors and two parasites in Egypt

Introduction: In past decades, leishmaniasis burden has been low across Egypt; however, changing environment and land use has placed several parts of the country at risk. As a consequence, leishmaniasis has become a particularly difficult health problem, both for local inhabitants and for multinational military personnel. Methods: To evaluate coarse-resolution aspects of the ecology of leishmaniasis transmission, collection records for sandflies and Leishmania species were obtained from diverse sources. To characterize environmental variation across the country, we used multitemporal Land Surface Temperature (LST) and Normalized Difference Vegetation Index (NDVI) data from the Moderate Resolution Imaging Spectroradiometer (MODIS) for 2005-2011. Ecological niche models were generated using MaxEnt, and results were analyzed using background similarity tests to assess whether associations among vectors and parasites (i.e., niche similarity) can be detected across broad geographic regions. Results: We found niche similarity only between one vector species and its corresponding parasite species (i.e., Phlebotomus papatasi with Leishmania major), suggesting that geographic ranges of zoonotic cutaneous leishmaniasis and its potential vector may overlap, but under distinct environmental associations. Other associations (e.g., P. sergenti with L. major) were not supported. Mapping suitable areas for each species suggested that northeastern Egypt is particularly at risk because both parasites have potential to circulate. Conclusions: Ecological niche modeling approaches can be used as a first-pass assessment of vector-parasite interactions, offering useful insights into constraints on the geography of transmission patterns of leishmaniasis.

Semantic relations extraction from unstructured information for domain ontologies enrichment

Based in internet growth, through semantic web, together with communication speed improvement and fast development of storage device sizes, data and information volume rises considerably every day. Because of this, in the last few years there has been a growing interest in structures for formal representation with suitable characteristics, such as the possibility to organize data and information, as well as the reuse of its contents aimed for the generation of new knowledge. Controlled Vocabulary, specifically Ontologies, present themselves in the lead as one of such structures of representation with high potential. Not only allow for data representation, as well as the reuse of such data for knowledge extraction, coupled with its subsequent storage through not so complex formalisms. However, for the purpose of assuring that ontology knowledge is always up to date, they need maintenance. Ontology Learning is an area which studies the details of update and maintenance of ontologies. It is worth noting that relevant literature already presents first results on automatic maintenance of ontologies, but still in a very early stage. Human-based processes are still the current way to update and maintain an ontology, which turns this into a cumbersome task. The generation of new knowledge aimed for ontology growth can be done based in Data Mining techniques, which is an area that studies techniques for data processing, pattern discovery and knowledge extraction in IT systems. This work aims at proposing a novel semi-automatic method for knowledge extraction from unstructured data sources, using Data Mining techniques, namely through pattern discovery, focused in improving the precision of concept and its semantic relations present in an ontology. In order to verify the applicability of the proposed method, a proof of concept was developed, presenting its results, which were applied in building and construction sector.

Learning domain-specific sentiment lexicons with applications to recommender systems

Search is now going beyond looking for factual information, and people wish to search for the opinions of others to help them in their own decision-making. Sentiment expressions or opinion expressions are used by users to express their opinion and embody important pieces of information, particularly in online commerce. The main problem that the present dissertation addresses is how to model text to find meaningful words that express a sentiment. In this context, I investigate the viability of automatically generating a sentiment lexicon for opinion retrieval and sentiment classification applications. For this research objective we propose to capture sentiment words that are derived from online users’ reviews. In this approach, we tackle a major challenge in sentiment analysis which is the detection of words that express subjective preference and domain-specific sentiment words such as jargon. To this aim we present a fully generative method that automatically learns a domain-specific lexicon and is fully independent of external sources. Sentiment lexicons can be applied in a broad set of applications, however popular recommendation algorithms have somehow been disconnected from sentiment analysis. Therefore, we present a study that explores the viability of applying sentiment analysis techniques to infer ratings in a recommendation algorithm. Furthermore, entities’ reputation is intrinsically associated with sentiment words that have a positive or negative relation with those entities. Hence, is provided a study that observes the viability of using a domain-specific lexicon to compute entities reputation. Finally, a recommendation system algorithm is improved with the use of sentiment-based ratings and entities reputation.