12828 research outputs found
Sort by
Tierschutz in der EU
No full textEingereicht von Jana MitterDiplomarbeit Johannes Kepler Universität Linz 2025Arbeit auf den öffentlichen PCs in den Bibliotheken der JKU+Medizin abrufba
International Journal of Mechanical Engineering Education / How orthographic projection engineering drawing support VHS students in their future career: An industrial perspective
No full textThis study explores the industrial perspective on the role of orthographic projection engineering drawing within Vocational High Schools (VHS), a representative secondary-level institution in Technical and Vocational Education and Training (TVET). The research examines how orthographic projection supports students’ preparedness for future careers in the automotive and mechanical industrial sectors. Conducted in 2024, the study employed qualitative design using semi-structured interviews with five experienced industrial practitioners selected through purposive sampling. These practitioners, working in prominent automotive and mechanical companies in Indonesia, provided insights into the skills expected of VHS graduates. Thematic analysis revealed three central themes: foundational technical skills, career readiness and skills application, and cognitive and analytical growth. Findings indicate that mastery of manual orthographic drawing significantly strengthens essential technical competencies, directly supports career-specific tasks, and enhances spatial reasoning and problem-solving abilities. The integration of manual drawing techniques with digital tools prepares students to meet the evolving demands of modern industry by fostering both adaptability and proficiency across analog and digital workflows.The Austria Ministry of Education, and Kemendikbud Dikti Indonesian Ministry of Education MPC-2023-0305
IEEE Transactions on Magnetics / On the vector potential formulation with an energy-based hysteresis model and its numerical solution
No full textDeutsche Forschungsgemeinschaft 492661287/TRR 361 10.55776/F9
Journal of Bodywork and Movement Therapies / Massage and exercise increase body awareness in healthy adults: A single blinded randomized controlled trial
No full textJournal of Geometry / Characterisation of quadratic spaces over the Hilbert field by means of the orthogonality relation
No full textAbstract An orthoset is a set equipped with a symmetric, irreflexive binary relation. With any (anisotropic) Hermitian space H , we may associate the orthoset ( P ( H ) , ⊥ ) , consisting of the set of one-dimensional subspaces of H and the usual orthogonality relation. ( P ( H ) , ⊥ ) determines H essentially uniquely.We characterise in this paper certain kinds of Hermitian spaces by imposing transitivity and minimality conditions on their associated orthosets. By gradually considering stricter conditions, we restrict the discussion to a narrower and narrower class of Hermitian spaces. Ultimately, our interest lies in quadratic spaces over countable subfields of R .A line of an orthoset is the orthoclosure of two distinct elements. For an orthoset to be line-symmetric means roughly that its automorphism group acts transitively both on the collection of all lines as well as on each single line. Line-symmetric orthosets turn out to be in correspondence with transitive Hermitian spaces. Furthermore, quadratic orthosets are defined similarly, but are required to possess, for each line "Equation missing" , a group of automorphisms acting on "Equation missing" transitively and commutatively. We show the correspondence of quadratic orthosets with transitive quadratic spaces over ordered fields. We finally specify those quadratic orthosets that are, in a natural sense, minimal: for a finite n ⩾ 4 , the orthoset ( P ( R n ) , ⊥ ) , where R is the Hilbert field, has the property of being embeddable into any other quadratic orthoset of rank n .Grantová Agentura České Republiky 20-09869L 25-20013L 25-20013L I 4579-N 10.55776/PIN5424624 10.55776/PIN542462
Journal of Population Economics / Life after loss: the causal effect of parental death on daughters’ fertility
No full textAbstract Using Austrian administrative data, this study examines the causal effect of parental death on daughters’ fertility through a difference-in-differences approach. The findings indicate that parental death leads to quantitatively insignificant changes in the number of children and the probability of childlessness. Complementary analyses show no substantial effects on labor market participation, residential mobility, or long-term mental health. The evidence suggests that fertility decisions remain largely unaffected by the logistical and emotional challenges of parental loss, highlighting the resilience of reproductive choices to external life shocks
ELECTROPHORESIS / Determination of Withanolides and Withanosides in Ashwagandha Based Products Using HPLC‐Drift‐Tube‐Ion‐Mobility Quadrupole Time‐of‐Flight Mass Spectrometry
No full textABSTRACT A total of 19 products based on extracts from Withania somnifera (L.) Dunal, better known by its more common name ashwagandha, and five products based on ashwagandha root powder were analyzed with respect to their content in the biologically relevant substances belonging to the group of withanolides and withanosides. Using HPLC coupled to drift‐tube ion‐mobility quadrupole time‐of‐flight mass spectrometry (DT‐IM‐QTOF‐MS), 19 withanolides and withanosides could be tentatively identified. The comparison of the results from the quantitative analysis with the information on the product labels showed that the percentage of withanolides and withanosides deviated from the stated specifications by at least a factor of two and at most a factor of 35
The Single-Mass Variable Flavor Number Scheme at Three-Loop Order : RISC Report Series, 25-04
No full textThe matching relations in the unpolarized and polarized variable flavor number scheme at three-loop order are presented in the single-mass case. They describe the process of massive quarks becoming light at large virtualities . In this framework, heavy-quark parton distributions can be defined. Numerical results are presented on the matching relations in the case of the single-mass variable flavor number scheme for the light parton, charm and bottom quark distributions. These relations are process independent. In the polarized case we generally work in the Larin scheme. To two-loop order we present the polarized massive OMEs also in the scheme. Fast numerical codes for the single-mass massive operator matrix elements are provided.Fonds zur Förderung der Wissenschaftlichen Forschung 10.55776/P2034
Using large language models and law-based rules over extracted knowledge graphs for the analysis of VAT chain-transaction cases in Austrian VAT law
No full textLegal decisions must be reliable and unambiguous, which is why they must be made on an explainable and logically founded basis. Yet, this is precisely where large language models (LLMs) still have their difficulties due to their tendency to hallucinate. As a result, the EU AI Act classifies AI applications in the legal domain as high-risk, requiring additional governance and safeguards. In this paper, we present a novel approach that combines the strengths of LLMs in processing natural-language text and the possibilities of rule-based decisions into a robust application that produces logic-based and law-based results in the context of VAT chain-transaction cases in Austrian VAT law. The framework required for the structured preparation of the decision basis is formed by a knowledge graph, which is also used to generate a graphical representation. The application was tested with a total of 167 practical cases and an overall accuracy of 94% in solving cases of VAT chain transactions was achieved. Despite the 6% inaccuracy, the findings of this work make a valuable contribution to the practical application of LLMs in the field of VAT chain transactions, as even the visual representation of a complex issue brings efficiency gains for solving a chain transaction. In addition to the automatic resolution of chain transactions, the visual representation is therefore a valuable artifact resulting from the developed application.Entscheidungen im Rechtsbereich müssen zuverlässig und eindeutig sein, weshalb sie auf einer erklärbaren und logisch begründeten Grundlage getroffen werden müssen. Genau damit haben aber Large Language Models (LLM) nach wie vor ihre Schwierigkeiten, und das ist mitunter ein Grund, warum AI-Applikationen im Rechtsbereich im AI-Act der EU als Hoch-Risiko-Applikationen eingestuft werden. Passend dazu stellen wir in dieser Arbeit einen neuen Ansatz vor, der die Stärken von LLMs in der Verarbeitung von natürlichsprachlichen Text und den Möglichkeiten von regelbasierten Entscheidungen zu einer robusten Anwendung zusammenführt. Die Anwendung liefert Ergebnisse im Zusammenhang mit umsatzsteuerlichen Reihengeschäften gemäß österreichischen Umsatzsteuergesetz, welche auf Logik und den geltenden Gesetzen basieren. Das benötigte Framework für die strukturierte Aufbereitung der Entscheidungsgrundlage wird durch einen Knowledge Graph gebildet, der gleichzeitig auch für die Erzeugung einer grafischen Darstellung genutzt wird. Insgesamt wurde die Anwendung mit 167 Fällen aus der Praxis getestet und dabei wurde eine Genauigkeit von 94% in der Lösung von umsatzsteuerlichen Reihengeschäften erzielt. Trotz der Ungenauigkeit von 6% leisten die Ergebnisse dieser Arbeit einen wertvollen Beitrag zur praktischen Anwendung von LLMs im Bereich der umsatzsteuerlichen Reihengeschäften, da bereits die visuelle Darstellung eines komplexen Sachverhalts eine Effizienzsteigerung bei der Lösung eines Reihengeschäfts mit sich bringt. Neben der automatischen Auflösung von Reihengeschäften ist daher die visuelle Darstellung ein wertvolles Artefakt, das aus der entwickelten Anwendung hervorgeht.Author Lukas Knogler, BScMasterarbeit Johannes Kepler Universität Linz 202
Eignung großer Sprachmodelle in Enterprise Content Management um PDF-Dokumente zugänglich und barriere-frei zu gestalten
No full textThis thesis examines the suitability of large language models (LLMs) for making PDF documents more accessible and barrier-free in enterprise content management (ECM). While accessibility is essential for inclusion and legal compliance, manual preparation of accessible PDFs remains slow and error-prone. The work integrates LLM-based methods into the Alfresco ECM platform, using few-shot learning, fine-tuning, and prompt engineering to generate accessibility-related annotations from decompressed PDF content streams.
The experiments demonstrate that LLMs can capture aspects of accessibility tagging, but the transformation results are not particularly strong when evaluated with standard NLP metrics. Instead, the main contribution lies in showing how such methods can be embedded in a real ECM workflow, and in identifying the limits of current LLMs for PDF-to-PDF accessibility translation. The thesis concludes that LLMs offer potential for future improvements, but significant challenges in syntax control, evaluation, and practical accuracy remain.Diese Masterarbeit untersucht die Eignung von Large Language Models (LLMs), um PDF- Dokumente im Rahmen von Enterprise Content Management (ECM) barriere-freier und zugänglicher zu machen. Obwohl Barrierefreiheit eine zentrale Voraussetzung für soziale Teilhabe und rechtliche Konformität ist, sind die derzeitigen manuellen Prozesse zur Erstellung barrierefreier PDFs aufwendig und fehleranfällig. Im Rahmen der Arbeit werden LLM-basierte Methoden in die Alfresco-ECM-Plattform integriert, wobei Few-Shot Learning, Fine-Tuning und Prompt Engineering eingesetzt werden, um Barrierefreiheits-Annotationen aus dekomprimierten PDF-Inhaltsströmen zu erzeugen.
Die Ergebnisse zeigen, dass LLMs zwar gewisse Aspekte des Taggings erfassen, die Transformationsresultate jedoch nach gängigen NLP-Metriken eher schwach ausfallen. Der wesentliche Beitrag der Arbeit liegt daher weniger im Nachweis hoher Modellgüte als vielmehr in der Einbettung solcher Ansätze in einen realen ECM-Workflow und in der Analyse der aktuellen Grenzen von LLMs bei der PDF-zu-PDF-Barrierefreiheitsumwandlung. Die Arbeit schließt mit dem Fazit, dass LLMs ein zukünftiges Potenzial besitzen, derzeit aber noch erhebliche Herausforderungen in Bezug auf Syntaxkontrolle, Evaluierung und praktische Genauigkeit bestehen.Author Jack HeseltineAbweichender Titel laut Übersetzung der Verfasserin/des VerfassersMasterarbeit Johannes Kepler Universität Linz 202