Search

We compared short-term physical performance adaptations to traditional strength training (TST) or complex strength training (CST) in a youth football (soccer) team during the preparatory period. A professional U19 male football team (n = 26; age = 17.3 ± 1.5 yrs) were divided in a TST (4 sets × 4 reps at 60–85% of one-repetition maximum [1RM] resistance training twice/week; n = 9), a CST (TST + plyometrics; n = 11) and a control (CON, only regular team’ practice; n = 6) group. The 1RM bench press, 1RM back squat, squat jump (SJ), countermovement jump (CMJ), Illinois agility test, 10-, 20- and 30-m sprint and Yo-Yo Intermittent Recovery Test Level 1 [Yo-Yo IR1] performance were tested before and after a 4-week preparatory period. Training load was quantified using the session rating of perceived exertion (s-RPE; RPE × duration) method. Both TST and CST groups improved 1RM-BP and 1RM-BS (d = 0.67–0.91). The CST group improved SJ, CMJ, 10-, 20- and 30-m sprint and Yo-Yo IR1 performance (d = 0.43–0.74). The CON group experienced moderate improvements in CMJ (d = 0.73). After adjustment for s-RPE, the TST improved Yo-Yo IR1 performance (d = 0.53), whereas all functional adaptations disappeared in the CST and CON group. No between-group differences were observed in s-RPE neither after gym nor field training. Adding TST or CST twice/week to regular pre-season U19 football training induced similar effects on maximal strength, albeit that CST seems more effective to improve functional capacity without being perceived as more demanding. Strength and conditioning coaches could implement CST to simultaneously enhance strength, power and intermittent exercise capacity in U19 football players.

The increasing prevalence of Open Science has brought reproducibility to the center of discussion of the scientific community as a requirement for ensuring the transparency and correctness of a research workflow. The current publishing landscape is evolving, as shown by the emergence of notebook technologies powering a new generation of interactive Web Journals. These use state-of-the-art interactive graphical visualizations and on-demand data processing to research papers, allowing readers to trace every step of the process, from raw data to the finalized visualization. Since there are many Research Notebook technologies and interactive graphical visualization solutions to choose from, we present a summary comparative overview of Web Journals and the Notebook engines that power the interactive, data driven visualizations inside their publications. Given our focus on visualization, our metrics are the support for the most advanced, popular and widely adopted data visualization frameworks. We conclude that Jupyter Notebook is currently the best alternative for the average user, given its popularity and support, combined with broad support for powerful and high-level interactive visualization grammars.

The current requirements for open data in the EU are increasing the awareness of researchers with respect to data management and data publication. Metadata is essential in research data management, namely on data discovery and reuse. Current practices tend to either leave metadata definition to researchers, or to assign their creation to curators. The former typically results in ad-hoc descriptors, while the latter follows standards but lacks specificity. In this exploratory study, we adopt a researcher-curator collaborative approach in five data publication cases, involving researchers in data description and discussing the use of both generic and domain-oriented metadata. The study shows that researchers working on familiar datasets can contribute effectively to the definition of metadata models, in addition to the actual metadata creation. The cases also provide preliminary evidence of cross-disciplinary descriptor use. Moreover, the interaction with curators highlights the advantages of data management, making researchers more open to participate in the corresponding tasks.

Over the past years, the amount of online offensive speech has been growing steadily. To successfully cope with it, machine learning are applied. However, ML-based techniques require sufficiently large annotated datasets. In the last years, different datasets were published, mainly for English. In this paper, we present a new dataset for Portuguese, which has not been in focus so far. The dataset is composed of 5,668 tweets. For its annotation, we defined two different schemes used by annotators with different levels of expertise. Firstly, non-experts annotated the tweets with binary labels (‘hate’ vs. ‘no-hate’). Secondly, expert annotators classified the tweets following a fine-grained hierarchical multiple label scheme with 81 hate speech categories in total. The inter-annotator agreement varied from category to category, which reflects the insight that some types of hate speech are more subtle than others and that their detection depends on personal perception. This hierarchical annotation scheme is the main contribution of the presented work, as it facilitates the identification of different types of hate speech and their intersections. To demonstrate the usefulness of our dataset, we carried a baseline classification experiment with pre-trained word embeddings and LSTM on the binary classified data, with a state-of-the-art outcome.

Research datasets include all kinds of objects, from web pages to sensor data, and originate in every domain. Concerns with data generated in large projects and well-funded research areas are centered on their exploration and analysis. For data in the long tail, the main issues are still how to get data visible, satisfactorily described, preserved, and searchable. Our work aims to promote data publication in research institutions, considering that researchers are the core stakeholders and need straightforward workflows, and that multi-disciplinary tools can be designed and adapted to specific areas with a reasonable effort. For small groups with interesting datasets but not much time or funding for data curation, we have to focus on engaging researchers in the process of preparing data for publication, while providing them with measurable outputs. In larger groups, solutions have to be customized to satisfy the requirements of more specific research contexts. We describe our experience at the University of Porto in two lines of enquiry. For the work with long-tail groups we propose general-purpose tools for data description and the interface to multi-disciplinary data repositories. For areas with larger projects and more specific requirements, namely wind infrastructure, sensor data from concrete structures and marine data, we define specialized workflows. In both cases, we present a preliminary evaluation of results and an estimate of the kind of effort required to keep the proposed infrastructures running. The tools available to researchers can be decisive for their commitment. We focus on data preparation, namely on dataset organization and metadata creation. For groups in the long tail, we propose Dendro, an open-source research data management platform, and explore automatic metadata creation with LabTablet, an electronic laboratory notebook. For groups demanding a domain-specific approach, our analysis has resulted in the development of models and applications to organize the data and support some of their use cases. Overall, we have adopted ontologies for metadata modeling, keeping in sight metadata dissemination as Linked Open Data.

The assessment of wind energy resource for the development of deep offshore wind plants requires the use of every possible source of data and, in many cases, includes data gathered at meteorological stations installed at islands, islets or even oil platforms—all structures that interfere with, and change, the flow characteristics. This work aims to contribute to the evaluation of such changes in the flow by developing a correction methodology and applying it to the case of Berlenga island, Portugal. The study is performed using computational fluid dynamic simulations (CFD) validated by wind tunnel tests. In order to simulate the incoming offshore flow with CFD models a wind profile, unknown a priori , was established using observations from two coastal wind stations and a power law wind profile was fitted to the existing data ( α =0.165). The results show that the resulting horizontal wind speed at 80 m above sea level is 16% lower than the wind speed at 80 m above the island for the dominant wind direction sector.

Research data are the cornerstone of science and their current fast rate of production is disquieting researchers. Adequate research data management strongly depends on accurate metadata records that capture the production context of the datasets, thus enabling data interpretation and reuse. This chapter reports on the authors' experience in the development of the metadata models, formalized as ontologies, for several research domains, involving members from small research teams in the overall process. This process is instantiated with four case studies: vehicle simulation; hydrogen production; biological oceanography and social sciences. The authors also present a data description workflow that includes a research data management platform, named Dendro, where researchers can prepare their datasets for further deposit in external data repositories.

Metadata puts research data in their context, making data intelligible and apt to sustain technology evolution and to be reused, in compliance with the FAIR principles. The workflow proposed in this work includes metadata generation in the context of research projects, created with the Dendro platform, and metadata originated in the interaction of people with the deposited data, created with the B2NOTE service from EUDAT. In our experiments, datasets are prepared with Dendro, taking into consideration general-purpose descriptors and domain-specific ones, then transparently deposited in B2SHARE. After publication, B2NOTE provides an environment where authors, other researchers, and any interested party can enrich the description with less formal comments, tags or keywords. This work contributes with (a) a set of use cases in several domains, (b) details on the descriptors used by authors in each case, and (c) reflections on the use of data after publication, using the B2NOTE contributions.

This chapter consists of a solution for the management of research data at a higher education and research institution. The chapter is based on a small-scale data audit study, which included contacts with researchers and yielded some preliminary requirements and use cases. These requirements led to the design of a data curation workflow involving the researcher, the curator, and a data repository. The authors describe the features of the data repository prototype, which is an extension to the widely used DSpace repository platform and introduced a set of features mentioned by the majority of the interviewed researchers as relevant for a data repository. The data repository platform contributes to the curation workflow at the university, with XML technology at its core—data is stored using XML documents, which can be systematically processed and queried unlike its original-format counterpart. This system is capable of indexing, querying, and retrieving, in whole or in part, datasets represented in tabular form. There is also the possibility of using elements from domain-specific XML schemas for the cataloguing process, improving the interoperability and quality of the deposited data.

UPData is a scientific data curation experiment currentlyunder development at University of Porto which aims todetermine the main digital preservation needs of several researchgroups at the university. In the course of the experiment,eight datasets have been collected from diversescientific domains. After conducting several interviews withresearchers working at U.Porto, we have concluded that fromtheir point of view, flexible data access is the most valuedcapability when analysing a preservation solution and thatoffering such access it is the best way to involve them inthe preservation workflow. We propose an extension to theDSpace repository platform to complement it with data curationcapabilities. In the proposed solution, the system ingestsExcel spreadsheets containing scientific data and translatesthem into XML documents which can then be queriedvia automatically generated XQuery statements. Researchersuse a search webpage designed for displaying deposited dataand applying various filters to it, retrieving the parts theyneed without having to scan each file. The collected datasetswill be used as test cases for data deposit, and also to evaluatethe effort required by the curation procedure.

A ciencia depende cada vez mais de dados e e reconhecida a dificuldade em armazenar, descrever e partilhar conjuntos de dados produzidos em ambiente de investigacao. Em muitos casos, os conjuntos de dados de investigacao apenas sao sujeitos a processos de gestao apos a publicacao dos resultados da sua analise. Esta abordagem a posteriori funciona bem para recursos estaticos, como por exemplo as publicacoes, mas nao resulta tao bem para recursos com contextos de producao dinâmicos, como e o caso dos conjuntos de dados. No caso destes tipo de recursos, o processo de gestao devera comecar no inicio das atividades de investigacao, tornando-se uma parte integrante do proprio processo cientifico. Verifica-se neste momento que os repositorios de dados cientificos dependem completamente dos curadores para efetuar a anotacao dos conjuntos de dados. Para facilitar a preparacao dos dados para o seu deposito, propomos um ambiente colaborativo de gestao de dados de investigacao, desenhado para ajudar pequenas equipas de investigacao a armazenar e descrever os seus conjuntos de dados, em preparacao para o seu posterior deposito. E suportado por dois componentes integrados: o UPBox?uma ?nuvem? privada desenhada para suportar o armazenamento dos ficheiros?e o DataNotes?uma solucao ?wiki? desenhada para os investigadores descreverem os seus ficheiros de forma colaborativa, construida sobre a Semantic MediaWiki. Os resultados preliminares de testes com utilizadores produziram respostas positivas as funcionalidades implementadas, o que permite concluir que estas respondem a necessidades reais dos investigadores