Search

Trusted Research Environments (TREs) enable the analysis of sensitive data under strict security assertions that protect the data with technical, organizational, and legal measures from (accidentally) being leaked outside the facility. While many TREs exist in Europe, little information is available publicly on the architecture and descriptions of their building blocks and their slight technical variations. To highlight on these problems, an overview of the existing, publicly described TREs and a bibliography linking to the system description are provided. Their technical characteristics, especially in commonalities and variations, are analysed, and insight is provided into their data type characteristics and availability. The literature study shows that 47 TREs worldwide provide access to sensitive data, of which two-thirds provide data predominantly via secure remote access. Statistical offices (SOs) make the majority of sensitive data records included in this study available.

Trust is an essential pre-condition for the acceptance of digital infrastructures and services. Transparency has been identified as one mechanism for increasing trustworthiness. Yet, it is difficult to assess to which extent and how exactly different aspects of transparency contribute to trust, or potentially impede it in cases of overwhelming complexity of the information provided. To address these issues, we performed two initial studies to help determining the factors that influence or have impact on trust, focusing on transparency across a range of elements associated with data, data infrastructures and virtual research environments. On one hand, we performed a survey among IT experts in the field of data science focusing on quality aspects in the context of re-using and sharing open source software, assessing issues such as the need for documentation, test cases, and accountability. On the other hand, we complemented this with a set of semi-structured interviews with senior researchers to address specific issues of the degree of transparency achievable with different approaches. They include, for example, the amount of transparency we can achieve with approaches from explainable AI, or the usefulness and limitations of data provenance in determining the suitability of data for reuse and others. Specifically, we consider mechanisms on three levels, i.e. technical, process-oriented as well as social mechanisms. Starting from attributes of trust in the “analogue world”, we aim to understand which of these can be applied in the digital world, how they differ, and what additional mechanisms need to be established, in order to support trust in complex socio-technological processes and their emergent results when the traditional approaches cannot be applied anymore.

Deep Learning has enabled significant progress towards more accurate predictions and is increasingly integrated into our everyday lives in real-world applications; this is true especially for Convolutional Neural Networks (CNNs) in the field of image analysis. Nevertheless, it has been shown that Deep Learning is vulnerable against well-crafted, small perturbations to the input, i.e., adversarial examples. Defending against such attacks is therefore crucial to ensure the proper functioning of these models—especially when autonomous decisions are taken in safety-critical applications, such as autonomous vehicles. In this work, shallow machine learning models, such as Logistic Regression and Support Vector Machine, are utilised as surrogates of a CNN based on the assumption that they would be differently affected by the minute modifications crafted for CNNs. We develop three detection strategies for adversarial examples by analysing differences in the prediction of the surrogate and the CNN model: namely, deviation in (i) the prediction, (ii) the distance of the predictions, and (iii) the confidence of the predictions. We consider three different feature spaces: raw images, extracted features, and the activations of the CNN model. Our evaluation shows that our methods achieve state-of-the-art performance compared to other approaches, such as Feature Squeezing, MagNet, PixelDefend, and Subset Scanning, on the MNIST, Fashion-MNIST, and CIFAR-10 datasets while being robust in the sense that they do not entirely fail against selected single attacks. Further, we evaluate our defence against an adaptive attacker in a grey-box setting.

Die Digitalisierung vollzieht sich auch in der Landwirtschaft in rasanter Geschwindigkeit. Die erheblichen Risiken bei der Anwendung von Lösungen, die in ihren Grundlagen und Auswirkungen oft nicht vollständig verstanden werden, bergen jedoch ein hohes Bedrohungspotenzial für die Resilienz und Nachhaltigkeit der Landwirtschaft. Dieser Artikel zeigt den Einsatz von Digitalen Zwillingen anhand von zwei ausgewählten Versuchsfarmen in Österreich. Das Projekt verfolgt das Ziel, durch den Aufbau modernster Versuchsfarmen als „Digitale Zwillinge“ eine zukunftsweisende Versuchsinfrastruktur für interdisziplinäre Forschung und Lehre auf internationalem Niveau zu etablieren. Es beinhaltet zudem ein Framework zur Integration von Anforderungen gemäß verschiedener Kriterien, um ein nachhaltiges Design von Digitalen Zwillingen in der Landwirtschaft zu ermöglichen.

Recordings of musical practices are kept in various public institutions and private depositories around the world. They constitute valuable data for ethnomusicological research and are substantial for the world's musical heritage. At the moment, there are no commonly used systems and standards for organizing, describing or categorizing these data, which makes their use difficult. In this paper, we discuss the required steps to make them findable, accessible, interoperable and reusable (FAIR), and outline action items to reach these goals. We show solutions that help researchers to manage their data over the whole research lifecycle and discuss the benefits of combining technologies from information science, music information retrieval, and linked data, with the aim of giving incentives for the ethnomusicology research community to actively participate in these developments in the future.

Data curation is a complex, multi-faceted task. While dedicated data stewards are starting to take care of these activities in close collaboration with researchers for many types of (usually file-based) data in many institutions, this is rarely yet the case for data held in relational databases. Beyond large-scale infrastructures hosting e.g. climate or genome data, researchers usually have to create, build and maintain their database, care about security patches, and feed data into it in order to use it in their research. Data curation, if at all, usually happens after a project is finished, when data may be exported for digital preservation into file repository systems. We present DBRepo, a semantic digital repository for relational databases in a private cloud setting designed to (1) host research data stored in relational databases right from the beginning of a research project, (2) provide separation of concerns, allowing the researchers to focus on the domain aspects of the data and their work while bringing in experts to handle classic data management tasks, (3) improve findability, accessibility and reusability by offering semantic mapping of metadata attributes, and (4) focus on reproducibility in dynamically evolving data by supporting versioning and precise identification/cite-ability for arbitrary subsets of data.

Meeting the conflicting goals of protecting and maintaining control over sensitive data while also allowing access by third parties constitutes a significant challenge. Secure data infrastructures support data visiting in a highly controlled and monitored environment which, if properly set-up and operated, provide high security guarantees through a combination of technical, legal and procedural mechanisms. To ease the process of deploying such a secure data infrastructure, we present a detailed documentation of the architecture and processes of such an infrastructure and provide a pre-configured reference implementation based entirely on open source software that can be flexibly configured to meet differing security requirements and deployment scenarios. We combine mechanisms for data visiting on secured infrastructure components with optional components of data anonymization and fingerprinting, covered by extensive logging and monitoring functions and embedded in defined processes and contractual frameworks. The set-up is based upon the experience of operating such a secure infrastructure in the medical domain for almost ten years, addressing the emerging need to make such a solution available to a larger set of stakeholders. We show that our system significantly enhances data visiting, offers a higher level of data isolation and present our open source reference implementation thereof.

Research Data Alliance Austria (RDA-AT) is a national RDA node dedicated to representing emerging research and data management communities throughout Austria. RDA-AT will operate as a formal participant of RDA Europe and RDA Global, linking Austrian data management initiatives and RDA Working and Interest Groups, providing assistance in adoption of RDA recommendations, and allowing Austrian stakeholders to benefit directly from RDA support mechanisms. These new connections will bring key Austrian issues to the global data management table, and global discussions back to Austria. This paper details the goals of the RDA-AT and describes its community and sustainability plans.

We generate synthetic data documenting COVID-19 cases in Austria by the means of an agent-based simulation model. The model simulates the transmission of the SARS-CoV-2 virus in a statistical replica of the population and reproduces typical patient pathways on an individual basis while simultaneously integrating historical data on the implementation and expiration of population-wide countermeasures. The resulting data semantically and statistically aligns with an official epidemiological case reporting data set and provides an easily accessible, consistent and augmented alternative. Our synthetic data set provides additional insight into the spread of the epidemic by synthesizing information that cannot be recorded in reality.

We present a novel method for interpolating univariate time series data. The proposed method combines multi-point fractional Brownian bridges, a genetic algorithm, and Takens’ theorem for reconstructing a phase space from univariate time series data. The basic idea is to first generate a population of different stochastically-interpolated time series data, and secondly, to use a genetic algorithm to find the pieces in the population which generate the smoothest reconstructed phase space trajectory. A smooth trajectory curve is hereby found to have a low variance of second derivatives along the curve. For simplicity, we refer to the developed method as PhaSpaSto-interpolation, which is an abbreviation for phase-space-trajectory-smoothing stochastic interpolation. The proposed approach is tested and validated with a univariate time series of the Lorenz system, five non-model data sets and compared to a cubic spline interpolation and a linear interpolation. We find that the criterion for smoothness guarantees low errors on known model and non-model data. Finally, we interpolate the discussed non-model data sets, and show the corresponding improved phase space portraits. The proposed method is useful for interpolating low-sampled time series data sets for, e.g., machine learning, regression analysis, or time series prediction approaches. Further, the results suggest that the variance of second derivatives along a given phase space trajectory is a valuable tool for phase space analysis of non-model time series data, and we expect it to be useful for future research.

The emergence of the World Wide Web facilitates the growth of user-generated texts in less-resourced languages. Sentiment analysis of these texts may serve as a key performance indicator of the quality of services delivered by companies and government institutions. The presence of user-generated texts is an opportunity for assisting managers and policy-makers. These texts are used to improve performance and increase the level of customers’ satisfaction. Because of this potential, sentiment analysis has been widely researched in the past few years. A plethora of approaches and tools have been developed—albeit predominantly for well-resourced languages such as English. Resources for less-resourced languages such as, in this paper, Amharic, are much less developed. As a result, it requires cost-effective approaches and massive amounts of annotated training data, calling for different approaches to be applied. This research investigates the performance of a combination of heterogeneous machine learning algorithms (base learners such as SVM, RF, and NB). These models in the framework are fused by a meta-learner (in this case, logistic regression) for Amharic sentiment classification. An annotated corpus is provided for evaluation of the classification framework. The proposed stacked approach applying SMOTE on TF-IDF characters (1,7) grams features has achieved an accuracy of 90%. The overall results of the meta-learner (i.e., stack ensemble) have revealed performance rise over the base learners with TF-IDF character n-grams.

Topic Modeling is a statistical process, which derives the latent themes from extensive collections of text. Three approaches to topic modeling exist, namely, unsupervised, semi-supervised and supervised. In this work, we develop a supervised topic model for an Amharic corpus. We also investigate the effect of stemming on topic detection on Term Frequency Inverse Document Frequency (TF-IDF) features, Latent Dirichlet Allocation (LDA) features and a combination of these two feature sets using four supervised machine learning tools, that is, Support Vector Machine (SVM), Naive Bayesian (NB), Logistic Regression (LR), and Neural Nets (NN). We evaluate our approach using an Amharic corpus of 14,751 documents of ten topic categories. Both qualitative and quantitative analysis of results show that our proposed supervised topic detection outperforms with an accuracy of 88% by SVM using state-of-the-art-approach TF-IDF word features with the application of the Synthetic Minority Over-sampling Technique (SMOTE) and with no stemming operation. The results show that text features with stemming slightly improve the performance of the topic classifier over features with no stemming.

Im November 2017 fand an der Universität Wien der RDA Europe Workshop „From Planning to Action. Towards the Establishment of an Austrian Research Infrastructure“ mit Unterstützung von RDA Europe, der Technischen Universität Wien und der Universität Wien statt. Es war der erste in Österreich abgehaltene Workshop in einer Reihe von europaweiten RDA Veranstaltungen. Ziel war es, über praktische Umsetzungen von RDA Empfehlungen, Implementierungsmaßnahmen der European Open Science Cloud und andere bedeutende Dateninitiativen zu diskutieren, um europäische Forschende und die Entwicklung von Infrastrukturen bestmöglich zu unterstützen. Die unterschiedlichen Themenbereiche bezogen sich auf transversale und disziplinspezifische Projekte, Datenzentren und Repositorien in Österreich. Die Vorträge reflektierten auch die europäische Perspektive von Forschungsdatenmanagement sowie politische und wissenschaftliche Initiativen von internationalen Organisationen. Ebenso wurden Empfehlungen von einzelnen Working Groups dargestellt und anhand von Projekten gezeigt, wie diese bereits lokal Anwendung finden. In der Veranstaltung wurde schließlich eine neu gegründete RDA Gruppe für Österreich angekündigt. Dieser RDA Austria Node wird, nach der formalen Konstituierungsphase, die österreichische Community innerhalb der neuen administrativen RDA-Strukturen repräsentieren.

The RDA Europe Workshop „From Planning to Action. Towards the Establishment of an Austrian Research Infrastructure“ was organized in Vienna in November 2017 with the support of RDA Europe, the Vienna University of Technology, and the University of Vienna. The workshop was the first RDA national event in Austria in a series of RDA Europe events dedicated to practical issues surrounding the adoption of RDA recommendations, the implementation of European Open Science Cloud and other key European data initiatives in order to support EU researchers and data centres. The different sections focused on transversal and domain-specific projects and infrastructures in Austria, such as data centres and data repositories that have emerged as a result of initiatives developed at the national level. The presentations also reflected the European perspective on research data management, focusing on the political and scientific work of international organisations. Furthermore there was an analysis of various RDA Europe working group recommendations and their potential for adoption in local contexts. The event also included the announcement of a new dedicated RDA group for Austria. This node will formally represent the Austrian data management community within the new RDA administrative structure as soon as formal procedures are completed.

Data management plans are free-form text documents describing the data used and produced in scientific experiments. The complexity of data-driven experiments requires precise descriptions of tools and datasets used in computations to enable their reproducibility and reuse. Data management plans fall short of these requirements. In this paper, we propose machine-actionable data management plans that cover the same themes as standard data management plans, but particular sections are filled with information obtained from existing tools. We present mapping of tools from the domains of digital preservation, reproducible research, open science, and data repositories to data management plan sections. Thus, we identify the requirements for a good solution and identify its limitations. We also propose a machine-actionable data model that enables information integration. The model uses ontologies and is based on existing standards.

In the era of research infrastructures and big data, sophisticated data management practices are becoming essential building blocks of successful science. Most practices follow a data-centric approach, which does not take into account the processes that created, analysed and presented the data. This fact limits the possibilities for reliable verification of results. Furthermore, it does not guarantee the reuse of research, which is one of the key aspects of credible data-driven science. For that reason, we propose the introduction of the new concept of Process Management Plans, which focus on the identification, description, sharing and preservation of the entire scientific processes. They enable verification and later reuse of result data and processes of scientific experiments. In this paper we describe the structure and explain the novelty of Process Management Plans by showing in what way they complement existing Data Management Plans. We also highlight key differences, major advantages, as well as references to tools and solutions that can facilitate the introduction of Process Management Plans.

Rescuing data from inaccessible or damaged storage media for the purpose of preserving the digital data for the long term is one of the dimensions of digital archaeology. With the current pace of technological development, any system can become obsolete in a matter of years and hence the data stored in a specific storage media might not be accessible anymore due to the unavailability of the system to access the media. In order to preserve digital records residing in such storage media, it is necessary to extract the data stored in those media by some means.One early storage medium for home computers in the 1980s was audio tape. The first home computer systems allowed the use of standard cassette players to record and replay data. Audio cassettes are more durable than old home computers when properly stored. Devices playing this medium (i.e. tape recorders) can be found in working condition or can be repaired, as they are usually made out of standard components. By re-engineering the format of the waveform and the file formats, the data on such media can then be extracted from a digitised audio stream and migrated to a non-obsolete format.In this paper we present a case study on extracting the data stored on an audio tape by an early home computer system, namely the Philips Videopac+ G7400. The original data formats were re-engineered and an application was written to support the migration of the data stored on tapes without using the original system. This eliminates the necessity of keeping an obsolete system alive for enabling access to the data on the storage media meant for this system. Two different methods to interpret the data and eliminate possible errors in the tape were implemented and evaluated on original tapes, which were recorded 20 years ago. Results show that with some error correction methods, parts of the tapes are still readable even without the original system. It also implies that it is easier to build solutions while original systems are still available in a working condition.

Interactive fiction and video games are part of our cultural heritage. As original systems cease to work because of hardware and media failures, methods to preserve obsolete video games for future generations have to be developed. The public interest in early video games is high, as exhibitions, regular magazines on the topic and newspaper articles demonstrate. Moreover, games considered to be classic are rereleased for new generations of gaming hardware. However, with the rapid development of new computer systems, the way games look and are played changes constantly. When trying to preserve console video games one faces problems of classified development documentation, legal aspects and extracting the contents from original media like cartridges with special hardware. Furthermore, special controllers and non-digital items are used to extend the gaming experience making it difficult to preserve the look and feel of console video games.This paper discusses strategies for the digital preservation of console video games. After a short overview of console video game systems, there follows an introduction to digital preservation and related work in common strategies for digital preservation and preserving interactive art. Then different preservation strategies are described with a specific focus on emulation. Finally a case study on console video game preservation is shown which uses the Planets preservation planning approach for evaluating preservation strategies in a documented decision-making process. Experiments are carried out to compare different emulators as well as other approaches, first for a single console video game system, then for different console systems of the same era and finally for systems of all eras. Comparison and discussion of results show that, while emulation works very well in principle for early console video games, various problems exist for the general use as a digital preservation alternative. We show what future work has to be done to tackle these problems.

Reproducibility and reusability of research results is an important concern in scientific communication and science policy. A foundational element of reproducibility and reusability is the open and persistently available presentation of research data. However, many common approaches for primary data publication in use today do not achieve sufficient long-term robustness, openness, accessibility or uniformity. Nor do they permit comprehensive exploitation by modern Web technologies. This has led to several authoritative studies recommending uniform direct citation of data archived in persistent repositories. Data are to be considered as first-class scholarly objects, and treated similarly in many ways to cited and archived scientific and scholarly literature. Here we briefly review the most current and widely agreed set of principle-based recommendations for scholarly data citation, the Joint Declaration of Data Citation Principles (JDDCP). We then present a framework for operationalizing the JDDCP; and a set of initial recommendations on identifier schemes, identifier resolution behavior, required metadata elements, and best practices for realizing programmatic machine actionability of cited data. The main target audience for the common implementation guidelines in this article consists of publishers, scholarly organizations, and persistent data repositories, including technical staff members in these organizations. But ordinary researchers can also benefit from these recommendations. The guidance provided here is intended to help achieve widespread, uniform human and machine accessibility of deposited data, in support of significantly improved verification, validation, reproducibility and re-use of scholarly/scientific data.