Hochschule Konstanz University of Applied Sciences
Hochschulschriftenserver der HTWG KonstanzNot a member yet
2319 research outputs found
Sort by
Make them Socialites: Supporting Social Entrepreneurs
No full textSocial entrepreneurs address pressing societal challenges, yet their efforts often remain underacknowledged, and the support they receive is frequently inadequate. This gap extends beyond practice and into academic research, where systematic insights into support mechanisms remain scarce. To help close this gap, this study offers a comprehensive analysis of the sources and types of support available to social entrepreneurs. Drawing on a systematic literature review of 39 academic articles, complemented by initial qualitative insights, the study identifies academic institutions, public agencies, incubators, and private actors such as investors as the main providers of support. However, one of the most influential sources of support for social venture development has been largely neglected in the literature. Across the reviewed studies, entrepreneurial education, training, and networking emerge as the most commonly provided forms of support. While practitioners confirm the relevance of networking, they highlight the critical value of more individualized support formats, such as coaching, which are less emphasized in the extant literature. Moreover, although social entrepreneurs face different challenges than commercial entrepreneurs, support mechanisms rarely reflect these distinct needs. The study concludes by highlighting the need for further research into the interdependencies between various support actors and the mechanisms through which effective support for social entrepreneurs is configured. ©2025 IEE
Development of a Research-Oriented Application for the Acquisition, Analysis, and Export of Sleep Metrics from Smartwatches
No full textAdvances in wearable technology have significantly enhanced the ability to monitor sleep in naturalistic, real-world environments, providing valuable insights beyond traditional laboratory-based methods. Smartwatches, equipped with sensors such as accelerom-eters, gyroscopes, and photoplethysmography (PPG), offer an accessible means to collect continuous sleep-related data. However, the limited access to raw sensor data and the use of proprietary algorithms in most commercial devices present substantial challenges for researchers aiming for transparency, reproducibility, and methodological flexibility. In response to this gap, this work introduces the development of a research-oriented software application specifically designed to enable the efficient extraction, visualization, and exportation of sleep metrics from smartwatches. The system empowers researchers to configure data acquisition parameters, access both processed metrics and raw sensor readings, and export data in customizable formats, such as CSV and JSON, thereby facilitating downstream scientific analysis. Particular attention was given to creating a user-friendly interface optimized for mobile devices, along with secure data handling mechanisms. This work highlights the growing importance of customizable, open-access tools in sleep research, offering a flexible alternative to closed commercial ecosystems. By bridging the gap between consumer devices and academic research needs, the proposed solution paves the way for broader adoption of wearable technology in decentralized sleep studies and fosters new possibilities for personalized health monitoring and longitudinal sleep assessment in diverse populations
ESG-Regelungen mit Bezug zur Wertschöpfungs- und Lieferkette
No full textDer Beitrag stellt die wichtigsten ESG-Regelungen mit Bezug zur Wertschöpfungs- und Lieferkette dar: LkSG, CSDDD, FLBR und EUDR. Er zeigt den aktuellen gesetzgeberischen Stand und die sich daraus ergebenden Implikationen für Unternehmen auf
Frederic Vester – Vordenker, Impulsgeber und Wegbereiter des Vernetzten Denkens und einer ökologisch-nachhaltigen Wirtschafts- und Lebensweise
No full textFür den Beitrag zu Vita, Werk und Wirkung haben die Autoren viele Details recherchiert und Materialien zusammengestellt, die Frederic Vesters Lebensweg anhand von eigenen Aussagen in Interviews, der Öffentlichkeit weniger bekannten Details und seiner bis heute einzigartigen, interdisziplinären Werke zum Vernetzten Denken und Handeln als Weg hin zur Nachhaltigkeit und Zukunftsfähigkeit illustrieren. Die kompakte Darstellung seiner verschiedenen Denkansätze und Werke in kursorischen Übersichten und Interpretationen, insbesondere auch hinsichtlich der zahlreich durchgeführten Systemstudien, erlaubt Einblicke in unterschiedliche Themenbereiche. Neben der Innovationskraft seiner Ansätze zeigen sie gleichzeitig seine wissenschaftliche und künstlerische Kreativität sowie seine Liebe zu den Menschen und zur Natur, die Frederic Vester zu einem außergewöhnlichen, interdisziplinären Wissenschaftler, insbesondere System- und Komplexitätsforscher, Didaktiker, Buch- und Filmautor sowie Unternehmens- und Politikberater von immer noch hoher Relevanz im Hinblick auf die aktuellen komplexen Herausforderungen machen
Recht und Governance
No full textUnternehmen stehen vor der Herausforderung, dass sie aus der Rechtsunsicherheit wegen bisher fehlender Regulierung im Umgang mit nicht personenbezogenen Daten nun ein reguliertes EU-Datenwirtschaftsrecht bewältigen müssen. Insbesondere kleinen und mittleren Unternehmen fehlt es oft an Data-Governance-Strukturen, um in diesem Umfeld effektiv und selbstbewusst Data Sharing zu betreiben, was ihre Bereitschaft zur Datenteilung stark hemmt. Der vorliegende Beitrag schlägt Grundsäulen einer rechtlich orientierten Data Governance vor, bestehend aus Dateninventarisierung (Daten als Asset), rechtlichen Schutzmöglichkeiten (IP-Management, insbesondere von Geschäftsgeheimnissen) und der vertraglichen Ausgestaltung der Datennutzung (Datenlizenzen)
Numerical Diffusion and its Impact on System-Identification for an Industrial Heating Process
No full textThis paper deals with system-identification for a distributed parameter heating process where a solid substrate is moving through a spatially extended heating zone and heated up by applying hot air to its surface. The temperature distribution inside the substrate is modeled in a spatial plane, where heat conduction is considered in the direction, perpendicular to the direction of movement. In contrast to previous work, where scalar model parameters (e.g.The thermal parameters of the substrate) have been identified, here, the quantities for the heat transfer (heat transfer coefficient and air temperature) are identified as functions yielding a significantly improved fit to the measurement data. This improved system-identification is performed for two early-lumping modeling approaches, which differ in the way the advection term in the governing Partial Differential Equation is discretized: one uses Eulerian coordinates, where the computational grid is stationary, whereas the second employs Lagrangian coordinates where the grid is moving with the substrate. The differences of the two approaches are discussed with the main focus on numerical diffusion. Especially its impact on the system-identification is investigated: Although the fit to the measurement is comparably good in both cases, very different solutions are obtained for the identified functions which, we argue, is due to the optimizer counteracting the smoothing effect of numerical diffusion
KI und Rhetorik: Trainingstool oder Langweiler?
No full textModeration: Kevin Schumacher, Schnitt: Thomas Heintz, Dauer: 1 h 5 min 7 sec, gedreht am 25.3.2025 am KIT Karlsruh
Optimizing Dataset Quality and Diversity in Neural Networks: A Study of the Vendi Score
No full textWe investigate methods for evaluating and optimizing dataset quality in the context of image-based object detection tasks, focusing on the Vendi Score (VS) and its enhanced variant, the quality-weighted Vendi Score (qVS).
These metrics provide a robust framework for assessing and improving dataset diversity, a critical factor in effectively training neural networks. In particular, the qVS allows for a customizable quality measure to account for dataset-specific characteristics.
In this paper, we introduce a specific weighting function based on the ratio of objects to total images to ensure balanced dataset composition and to emphasize the importance of diverse object representations. Feature extraction techniques are employed to represent image objects in a unified feature space, enabling similarity calculations and diversity assessments. Additionally, a multi-objective genetic algorithm (MOGA) is utilized to optimize datasets across multiple classes, maximizing diversity while maintaining class balance. Our experimental results demonstrate that models trained on subsets optimized for diversity using the qVS achieve improved performance. Diverse subsets lead to higher precision and generalization capabilities compared to randomly selected or less diverse datasets of the same size. Especially when sampling minibatches during training this method may prove beneficial, as it allows for a more representative sample of the dataset.
These findings underscore the pivotal role of dataset diversity in enhancing neural network performance and highlight the utility of the VS and qVS as valuable tools for strategically shaping dataset composition to improve outcomes