Mining biomedical information from scientific literature

The rapid evolution and proliferation of a world-wide computerized network, the Internet, resulted in an overwhelming and constantly growing amount of publicly available data and information, a fact that was also verified in biomedicine. However, the lack of structure of textual data inhibits its direct processing by computational solutions. Information extraction is the task of text mining that intends to automatically collect information from unstructured text data sources. The goal of the work described in this thesis was to build innovative solutions for biomedical information extraction from scientific literature, through the development of simple software artifacts for developers and biocurators, delivering more accurate, usable and faster results. We started by tackling named entity recognition - a crucial initial task - with the development of Gimli, a machine-learning-based solution that follows an incremental approach to optimize extracted linguistic characteristics for each concept type. Afterwards, Totum was built to harmonize concept names provided by heterogeneous systems, delivering a robust solution with improved performance results. Such approach takes advantage of heterogenous corpora to deliver cross-corpus harmonization that is not constrained to specific characteristics. Since previous solutions do not provide links to knowledge bases, Neji was built to streamline the development of complex and custom solutions for biomedical concept name recognition and normalization. This was achieved through a modular and flexible framework focused on speed and performance, integrating a large amount of processing modules optimized for the biomedical domain. To offer on-demand heterogenous biomedical concept identification, we developed BeCAS, a web application, service and widget. We also tackled relation mining by developing TrigNER, a machine-learning-based solution for biomedical event trigger recognition, which applies an automatic algorithm to obtain the best linguistic features and model parameters for each event type. Finally, in order to assist biocurators, Egas was developed to support rapid, interactive and real-time collaborative curation of biomedical documents, through manual and automatic in-line annotation of concepts and relations. Overall, the research work presented in this thesis contributed to a more accurate update of current biomedical knowledge bases, towards improved hypothesis generation and knowledge discovery.

Um modelo de medição e gestão de desempenho para as universidades públicas

Nas últimas décadas as instituições de ensino superior têm sido alvo de uma pressão crescente para aumentar a sua eficiência e a sua eficácia. Fatores como a diversificação da oferta formativa, a massificação, o aumento da internacionalização, entre outros, conduziram a uma maior monitorização das instituições e, por sua vez, geraram o desenvolvimento de novos sistemas de informação. A constante solicitação de informação, quer pelo Estado, quer pelo mercado constitui a base para a definição do objetivo desta investigação: construir um modelo integrado de medição e gestão de desempenho para as universidades públicas e testá-lo no universo português. Para a construção conceptual deste modelo foi realizada uma revisão da literatura baseada em diferentes contextos: organizacional, europeu e nacional. Este modelo foi depois aplicado às universidades públicas portuguesas recorrendo a dados disponibilizados por entidades públicas, tendo em conta a área de educação e formação (CNAEF). Uma análise descritiva aos dados constituiu uma contribuição desta investigação para a prática, no sentido de que permitiu fornecer um conjunto de recomendações às universidades e às entidades oficiais responsáveis pela recolha de dados a nível nacional, relativamente aos sistemas de informação e processos de recolha de dados. O modelo proposto constitui a contribuição teórica desta investigação, no sentido de que integra, no mesmo modelo, as diferentes dimensões de desempenho, a visão dos diferentes stakeholders, quer no contexto do ensino, quer no da investigação, quer no da própria gestão da universidade. A vertente analítica deste modelo, representada pelas diferentes relações entre os grupos de indicadores, foi testada recorrendo à técnica de análise de equações estruturais baseada na variância (nomeadamente Partial Least Squares) em quatro áreas CNAEF. Os resultados demonstraram que o comportamento em termos de medição e gestão de desempenho difere consoante a CNAEF, identificando um maior ajustamento às áreas hard e evidenciando que as áreas soft necessitam de indicadores mais ajustados às suas especificidades.

Integrated coastal geoengineering approach for maritime environments

This dissertation introduces several methodological approaches which integrate a proposed coastal management model in an interdisciplinary perspective. The research presented herein is displayed as a set of publications comprising different thematic outlooks. The thesis develops an integrated coastal geoengineering approach which is intrinsically linked to the studied maritime environments. From sandy coasts and marine works to rocky platforms and sea cliffs, this study includes field work between Caminha – Figueira da Foz (NW Portugal) and Galicia (NW Spain). The research also involves an analysis and geological-geotechnical characterisation of natural rock (armourstone) and artificial units (concrete blocks) applied to coastal structures. The main goal is to contribute to the characterisation and re-evaluation of georesources and to determine armourstone suitability and availability from its source (quarry). It was also important to diagnose the geomaterials in situ concerning their degradation/deterioration level on the basis of the current status of the coastal protection works in order to facilitate more efficient monitoring and maintenance, with economic benefits. In the rocky coast approach the coastal blocks were studied along the platform, but also the geoforms were studied from a coastal morphodynamics point of view. A shoreline evolution analysis was developed for sandy coasts through Digital Shoreline Analysis System (DSAS) extension. In addition, the spatial and statistical analysis applied to sea cliffs allowed the establishment of susceptibility zones to erosion and hazardous areas. All of these studies have different purposes and results however, there is a common denominator – GIS mapping. Hence, apart from the studied coastal environment, there is an integrated system which includes a sequence of procedures and methodologies that persisted during the research period. This is a step forward in the study of different coastal environments by using almost the same methodologies. This will allow the characterisation, monitoring and assessment of coastal protection works, rocky coasts, and shore platforms. With such data, it is possible to propose or recommend strategies for coastal and shoreline management based on several justifications in terms of social, economic, and environmental questions, or even provide a GIS-based planning support system reinforced by geocartographic decisions. Overall the development of the applied cartography embraces six stages which will allow the production of detailed maps of the maritime environment: (1) high-resolution aerial imagery surveys; (2) visual inspection and systematic monitoring; (3) applied field datasheet; (4) in situ evaluation; (5) scanline surveying; and (6) GIS mapping. This thesis covers fundamental matters that were developed over the course of scientific publication and as a consequence they represent the results obtained and discussed. The subjects directly related to the thesis architecture are: (i) cartography applied to coastal dynamics (including an art historical analysis as a tool to comprehend the coastal evolution and the littoral zone); (ii) georesources assessment (the role of cartography in georesources zoning, assessment and armourstone durability); (iii) coastal geoengineering applications and monitoring (Espinho pilot site in NW Portugal as an experimental field); (iv) rocky coast and shore platform studies and characterisation; (v) sandy and mixed environment approaches; (vi) coastal geosciences GIS mapping and photogrammetric surveying (coastal geoengineering); and (vii) shoreline change mapping and coastal management strategies (the CartGalicia Project as an example – NW Spain). Finally, all of these thematic areas were crucial to generate the conceptual models proposed and to shape the future of integrated coastal coastal geoengineering management.