Digital Library of Gesellschaft für Informatik e.V.
Not a member yet
40758 research outputs found
Sort by
Open Data im Spiegel der wissenschaftlichen Literatur – Eine quantitative Literaturanalyse von 2005–2024
No full textIn den letzten zwanzig Jahren hat sich Open Data zu einem zentralen Forschungsbereich entwickelt, der eine Vielzahl von wissenschaftlichen Disziplinen berührt. Durch Analyse von 13.538 wissenschaftlichen Literaturquellen aus der EBSCOhost-Datenbank werden die Eigenschaften dieses Forschungsfelds transparent gemacht. Dabei werden zentrale Themenfelder, theoretische Ansätze, eingesetzte Forschungsmethoden und Entwicklungstrends identifiziert. Um dies zu erreichen, wurden unterschiedliche Verfahren und Anwendungssysteme zur Extraktion und Analyse bibliografischer Daten angewendet. Die Untersuchung zeigt, dass das Forschungsgebiet Open Data durch eine ausgeprägte thematische und methodologische Vielfalt gekennzeichnet ist. Thematisch erstreckt sich die Forschung von Fragen der staatlichen Transparenz über wirtschaftliche Anwendungen bis hin zu informationstechnischen Innovationen bei der Strukturierung und Integration offener Datenbestände. Dabei kommen Theorien aus unterschiedlichen Disziplinen zur Anwendung, wobei neben Ökonomie und Informatik auch psychologische und politikwissenschaftliche Theorieansätze zu verzeichnen sind. Methodologisch finden sowohl qualitative als auch quantitative Ansätze Anwendung, die von Fallstudien über Experimente bis hin zur Prototypenentwicklung reichen. Die Literaturanalyse deutet darauf hin, dass das Forschungsfeld Open Data künftig durch moderne Analysemethoden – insbesondere aus dem Umfeld der generativen KI – stimuliert wird. Over the last twenty years, open data has developed into a central field of research that touches on a variety of scientific disciplines. By analyzing 13,538 scientific literature sources from the EBSCOhost database, the characteristics of this field of research are made transparent. Central topics, theoretical approaches, research methods used and development trends are identified. To achieve this, various methods and application systems were used to extract and analyze bibliographic data. The study shows that the research field of open data is characterized by a pronounced thematic and methodological diversity. Thematically, the research ranges from government transparency and economic applications to information technology innovations in the structuring and integration of open data sets. Theories from various disciplines are applied, with psychological and political science theories being used alongside theories from economics and computer science. Methodologically, both qualitative and quantitative approaches are used, ranging from case studies and experiments to prototype development. The literature analysis indicates that the research field of open data will be stimulated in the future by modern analysis methods—particularly from the field of generative AI
From Effort Reduction to Effort Management: An Expectancy Theory Perspective on Professionals’ Work Practices with Generative AI
No full textGenerative Artificial Intelligence (GenAI) is adopted by knowledge workers to boost productivity, yet its specific characteristics such as probabilistic outputs and human-level content generation may change how professionals think about their effort. Prior literature has warned about unintended side effects of AI, but experiments on effort reduction when working with AI – which could threaten performance – reported mixed results. GenAI’s rapid adoption combined with its specific characteristics make it critical and timely to clarify how GenAI influences knowledge workers’ effort in professional settings. The qualitative study draws on 21 interviews with knowledge workers who frequently use GenAI for work. A directed content analysis, guided by expectancy theory and social loafing frameworks, revealed that most interviewees do not simply reduce effort, but rather strategically reallocate or even increase effort. They continuously learn to steer GenAI, viewing themselves as process administrators. The traditional group-based mechanisms of reduced effort or diffused responsibility do not seem to be directly transferable to human–GenAI dyads in professional settings. By revealing that GenAI reshapes the factors that influence effort rather than simply eroding motivation, providing a multifaceted view of effort investment beyond mere reduction, and highlighting the interplay between human relationships and GenAI-facilitated work, this research advances the discourse on human-(Gen)AI dynamics and the unintended consequences of (Gen)AI. Recognizing these shifts when setting policies and expectations enables organizations to benefit from GenAI’s potential while mitigating potential risks to performance
Restoring Capacity at Work: Evaluating a Virtual Nature Experience for Public Administration Employees
No full textPublic administration employees face mounting cognitive demands that negatively impact well-being and performance. This study investigates the effect of a virtual realtiy nature simulation as a resource improving employee well-being, utilizing Job-Demand Resources Model and Attention Restoration Theory. To this end, we test a sample of public administration employees in a quasi-experimental pre-post design to examine the effects of a VR nature simulation in contrast with traditional breaktime and the effect on attention restoration and perceived well-being. We hypothesize a stronger positive effect on attention restoration in the VR group, increasing subjective well-being. Our study can thus provide an innovative low-threshold immersive technology to strengthen resilience in public administration workplaces
SkyFusion: A Physics-Informed GAN-Transformer Framework for Intra-Hour Sky Forecasting and PV Prediction
No full textReliable integration of renewable energy sources into electrical grids hinges on accurate forecasting of sky conditions and photovoltaic (PV) power generation [Nie23]. However, existing
methods, whether traditional numerical weather prediction or current deep learning models, have
faced challenges in reliably generating realistic intra-hour sky conditions (e.g., at 15-minute horizons) and associated PV power forecasts, primarily due to the complex and stochastic dynamics of cloud motion [Nie23]. In this work, we address this gap, demonstrating successful probabilistic sky forecasting and associated PV predictions at a 15-minute horizon. To address prior limitations, we introduce SkyFusion, a novel, modular pipeline that synergistically combines physics-informed latent-space modeling (PhyCell) [LG20], transformer-based autoregression, and a multi-stage
generative adversarial network (GAN) approach. Although GAN components faced initial stability
challenges, clear strategies for future integration are outlined, paving the way towards high-fidelity, physically plausible intra-hour sky sequences and robust probabilistic PV power prediction
WSRE 2026 — Call for Papers — 28. Workshop Software-Reengineering & -Evolution WSRE 2026
No full textCall for Papers für den 28. Workshop Software-Reengineering & -Evolution WSRE 2026. Unser Ziel ist die Förderung der Zusammenarbeit und der Austausch zwischen Forschung und Praxis im deutschsprachigen Raum zu den Themen Software-Reengineering, Software-Wartung und Software-Evolution. Darunter verstehen wir prinzipiell alle Aktivitäten rund um die Analyse, Bewertung, Visualisierung, Verbesserung, Migration und Weiterentwicklung von Software-Systemen. Wir laden Forscher*innen und Praktiker*innen herzlich ein, beim WSRE über Erfahrungen, Projekte, Forschungsergebnisse, Methoden und Werkzeuge in diesem Bereich zu berichten, ihre aktuellen Arbeiten vorzustellen und in einem offenen Umfeld konstruktiv zu diskutieren
Modeling and programming according to the sentence structure of natural languages
No full textIn this essay we describe how the grammar of modelling and implementing of sociotechnical systems, today mainly based on object-oriented methods, is enhanced by subjects. We want to add the subject to the grammar of programming as it is in natural languages. People find it easier to describe sociotechnical systems because they can use the same grammatical patterns as in their daily communication
SHAPing Latent Spaces in Facial Attribute Classification Models
No full textThis study investigates the use of SHAP (SHapley Additive exPlanations) values as an explainable artificial intelligence (xAI) technique applied on a facial attribute classification task. We analyse the consistency of SHAP value distributions across diverse classifier architectures that share the same feature extractor, revealing that key features driving attribute classification remain stable regardless of classifier architecture. Our findings highlight the challenges in interpreting SHAP values at the individual sample level, as their reliability depends on the model’s ability to learn distinct class-specific features; models exploiting inter-class correlations yield less representative SHAP explanations. Furthermore, pixel-level SHAP analysis reveals that superior classification accuracy does not necessarily equate to meaningful semantic understanding; notably, despite FaceNet exhibiting lower performance than CLIP, it demonstrated a more nuanced grasp of the underlying class attributes. Finally, we address the computational scalability of SHAP, demonstrating that KernelExplainer becomes infeasible for high-dimensional pixel data, whereas DeepExplainer and GradientExplainer offer more practical alternatives with trade-offs. Our results suggest that SHAP is most effective for small to medium feature sets or tabular data, providing interpretable and computationally manageable explanations
Wie können Automatisierung und Künstliche Intelligenz die Justiz bestmöglich unterstützen?
No full textDie Digitalisierung in der Justiz ermöglicht perspektivisch in zunehmendem Maße die Nutzung von Automatisierung und Künstlicher Intelligenz (KI), um Abläufe und Geschäftsprozesse in Justizbehörden zu vereinfachen und zu beschleunigen. Damit verbunden sind aus Sicht der Praxis und aller Stakeholder jedoch noch viele Hürden und offene Fragen – darunter die nach der Auswahl und Priorisierung geeigneter Anwendungsfälle und solche zur erfolgreichen Erprobung und gewinnbringenden Nutzung derartiger technologischer Möglichkeiten. Mit Hilfe etablierter Methoden aus dem Innovationsmanagement für Unternehmen und inspiriert durch den Legal Design-Ansatz haben wir die Bedarfe am Einsatz von Automatisierung und KI in der Justiz in Sachsen systematisch erfasst, mit entsprechenden Kennzahlen versehen und ausgewertet. Die Methodik sowie die Ergebnisse präsentieren wir in diesem Beitrag