Offenburg University of Applied Sciences
Hochschulschriftenserver der Hochschule OffenburgNot a member yet
6641 research outputs found
Sort by
RoBuddy – An Innovative Research Project on AI in Office Environments
No full textIn the rapidly evolving field of artificial intelligence (AI), the integration of AI-powered systems in office environments holds significant transformative potential. RoBuddy, an innovative research project, explores the application of AI in enhancing workplace efficiency and functionality. The project examines the role of AI in automating routine tasks, fostering creativity, and promoting strategic focus among employees.
A key focus of RoBuddy is the incorporation of positive emotional experiences in AI-human interactions, recognizing their critical impact on performance and motivation. By addressing the emotional dimensions of AI, the system offers a user-friendly and engaging interface through its innovative avatar designs, ensuring adaptability and relatability within diverse office settings.
This study also considers the implications of the EU AI Act, evaluating ethical and regulatory aspects of AI deployment in workplaces. Initial findings from questionnaires and visualizations reveal employees’ preferences and perceptions regarding digital assistants, providing valuable insights into user-centered design.
RoBuddy is a new approach on the harmonious integration of advanced technology with human-centric design, contributing to the broader discourse on AI’s role in redefining modern office dynamics. The project underscores the importance of aligning technological innovation with emotional intelligence to create systems that are both efficient and empathetic
Designing AI avatars for skill assessment: enhancing accessibility and trustworthiness
No full textThe collaborative research project KISEE (AI-based Skill Assessment and Development) explores the design possibilities of an interactive AI avatar to enhance access to future skills, particularly for individuals from educationally disadvantaged backgrounds. The avatar aims not only to complement traditional counseling approaches, such as chatbots or employment agencies, but also to make them more accessible, scalable, and effective. By providing personalized guidance, the avatar assesses existing competencies for identifying new career opportunities and facilitates access to training programs and certifications.
As part of the KISEE project, a systematic literature review of 33 academic studies was conducted to identify key factors influencing avatar design. The findings highlight the following aspects:
Gender and Stereotypes:
Avatars are subject to gendered perceptions. While female avatars are often associated with assistant roles, masculine avatars can increase interaction intensity. A gender-neutral design might help avoid stereotypes, but users tend to unconsciously assign gender based on contextual cues such as conversation topics or linguistic features.
Avatar Design and the Uncanny Valley:
Striking a balance between realism and stylization is crucial to avoiding the uncanny valley effect. Human-like avatars are generally perceived as more trustworthy, whereas stylized avatars elicit more positive emotional responses and may reduce cognitive load.
Trustworthiness:
The acceptance of an AI avatar strongly depends on transparency and clearly defined roles. Users prefer avatars in supportive rather than social roles and generally trust real humans more. A carefully balanced design—realistic yet visibly artificial—can maximize both trust and acceptance.
These findings provide valuable insights into designing effective, trustworthy, and inclusive AI avatars that promote equitable access to future skills and sustainable professional development
Pflichten zur Vertragsgestaltung unter DORA für IT-Dienstleister
No full textDie EU-Verordnung Nr. 2022/2554 über die digitale operationale Resilienz im Finanzsektor führt zwar einen wenig einprägsamen Namen, dafür aber dank der englischsprachigen Fassung eine griffige Abkürzung: DORA (Digital Operational Resilience Act). Erlassen am 14.12.2022 (ABl. v. 27.12.2022 Nr. L 333, S. 1) tritt sie am 17.1.2025 in Kraft (Art. 64 DORA), was angesichts direkt oder auch nur mittelbar zuliefern. Dass diese EU-Verordnung auch unmittelbare Pflichten für diese bereithält, wird manchem zuweilen erst jetzt klar
From Dashboards to Dialogues: Evaluating the Impact of a Conversational Interface on a Business Intelligence Platform
No full textThe study examines how embedding a conversational chatbot into a KPI-driven business intelligence (BI) platform affects executive decision-making in marketing and sales. Using a quantitative survey of marketing and sales executives, it finds that chatbot-enhanced interaction increases perceived system value by improving performance expectancy (via higher perceived output quality and time savings) and effort expectancy (via reduced cognitive load and greater computer self-efficacy). Overall, the results suggest that conversational BI can make analytics more accessible and efficient, strengthening executives’ decision confidence and offering actionable design implications for decision-support systems
Runners with lower dynamic stability exhibit better running economy: Results from a randomized crossover study of trained runners across various running speeds and footwear conditions
No full textBackground
Dynamic stability has been proposed as a factor influencing running economy, but the nature and strength of this relationship remain poorly understood. Further, advanced footwear technology (AFT) has been widely adopted as it improves running economy, but it may also compromise dynamic stability due to its compliant midsole. Understanding the relationship between dynamic stability and running economy and how both can be affected through footwear could have important implications for performance optimization and footwear design.
Methods
21 trained runners completed treadmill trials in four AFT models and their habitual SFT shoes at three individualized speeds. Local dynamic stability (LDS) was quantified via maximum Lyapunov exponents from seven segmental angular velocities, aggregated per trial. Running economy was measured as cost of transport. Linear mixed-effects models were used to assess relationships between LDS, COT, and footwear.
Results
Aggregated LDS was negatively associated with COT (p = 0.036), indicating that runners with lower dynamic stability had better running economy. Further, footwear condition had no significant effect on LDS (p = 0.060–0.359), suggesting that AFT does not compromise running stability compared to habitual SFT running shoes.
Conclusions
This study is the first to demonstrate that an aggregate of LDS across multiple body segments is negatively associated with COT, indicating that runners with lower dynamic stability exhibit better running economy. AFT did not affect LDS compared to habitual SFT, suggesting that alterations of footwear characteristics within commercially available models have limited effects on running stability
§ 110a Elektronische Aktenführung; Verordnungsermächtigungen
No full textDieser umfassende Kommentar zum OWiG orientiert sich an den in der Praxis entscheidungserheblichen Fragen und erörtert diese auf wissenschaftlichem Niveau in prägnanter und leicht verständlicher Weise. Er bietet auch dort Lösungsvorschläge an, wo Gerichtsentscheidungen bisher noch nicht vorliegen.
Aufgezeigt werden dabei auch Querverbindungen zu benachbarten Rechtsgebieten, vor allem zum Straf- und Strafprozessrecht.
Den Bedürfnissen der Benutzer entsprechend enthält der Anhang u.a. eine Reihe bundesrechtlicher Regelungen, auf die das OWiG verweist oder an die es anknüpft, sowie eine größere Anzahl landesrechtlicher Zuständigkeitsvorschriften
Integrating Cyber Threat Intelligence with SIEM for Enhanced Defensive Cybersecurity
No full textAs cyber threats continue to evolve at an alarming pace, adopting proactive cybersecurity strategies has become more crucial than ever. Traditional Security Information and Event Management (SIEM) systems are typically designed to detect internal threats but often fall short when it comes to providing real-time intelligence on external adversaries. This thesis examines how integrating Cyber Threat Intelligence (CTI) with SIEM systems can strengthen threat detection, reduce false positives, and improve incident response capabilities.
The research focuses on automating the ingestion of Indicators of Compromise (IOCs) from sources like ThreatFox, OpenPhish, and AlienVault OTX. The collected data is then structured and correlated within Splunk to identify security threats in real-time. Through this implementation, the system successfully detected malicious domains, phishing campaigns, and command-and-control (C2) infrastructures, significantly enhancing the efficiency of security monitoring.
The findings indicate that combining CTI with SIEM improves threat visibility and detection accuracy while reducing the dependence on reactive security models. However, challenges such as inconsistent data quality, manual IOC ingestion, and the complexities of integrating different systems were identified. Future advancements should prioritize automated threat intelligence processing, AI-driven threat correlation, and real-time adaptive security measures to further enhance cybersecurity defenses
Toolset zur niederschwelligen Partizipation der mittelständischen Industrie am Energiemarkt der Zukunft
No full textDer aktuelle wissenschaftlich-technische Stand beschreibt industrielle Energieflexibilität als die Fähigkeit, Produktionsprozesse schnell und kosteneffizient an volatile Energiemärkte anzupassen, um etwa Lastspitzen zu reduzieren oder Preisvorteile zu nutzen. Ziel der Untersuchung im Projekt FlexGUIde war die Entwicklung und Validierung datenbasierter Tools, die KMU befähigen, Flexibilitätspotenziale zu identifizieren, wirtschaftlich zu bewerten und praktisch zu nutzen. Methodisch kombinierte das Projekt reale Betriebsdaten mit KI-gestützten Prognoseverfahren, modellprädiktiver Steuerung und ökonomischen Analysen, die in mehreren Reallaboren aus der mittelständischen Industrie getestet wurden. Die Ergebnisse zeigen, dass sich Lastspitzen signifikant reduzieren und Energiekosteneinsparungen zwischen 5 % und 15 % im Zielsegment von Kunden mit einem Jahresverbrauch > 10 GWh erzielen lassen, sofern Flexibilitätsmaßnahmen niedrigschwellig integriert werden. Daraus folgt, dass datenbasierte, interoperable Systeme ein entscheidender Hebel für Effizienzsteigerung und Kostenreduktion in der Industrie sind und künftig eine zentrale Rolle bei der Integration von Energiemanagement und Marktpartizipation spielen können.The current state of scientific and technical knowledge defines industrial energy flexibility as the ability to rapidly and cost-effectively adapt production processes to volatile energy markets — for example, to reduce peak loads or take advantage of price fluctuations. The objective of the FlexGUIde project was to develop and validate data-driven tools that enable SMEs to identify, economically evaluate, and practically exploit their flexibility potentials. Methodologically, the project combined real operational data with AI-based forecasting methods, model-predictive control, and economic analyses, all of which were tested in several real-world industrial laboratories. The results show that peak loads can be significantly reduced and energy cost savings of between 5 % and 15 % can be achieved in the target segment of customers with an annual consumption above 10 GWh — provided flexibility measures are integrated with low implementation barriers. Consequently, data-driven, interoperable systems represent a key lever for improving efficiency and reducing costs in industry and are expected to play a central role in the future integration of energy management and market participation
DRIP: DRop unImportant data Points - Enhancing Machine Learning Efficiency with Grad-CAM-Based Streaming Data Prioritization for On-Device Training
No full textSelecting data points for model training is critical in machine learning. Effective selection methods can reduce the labeling effort, optimize on-device training for embedded systems with limited data storage, and enhance the model performance. This paper introduces a novel algorithm that uses Grad-CAM to make online decisions about retaining or discarding data points. Optimized for embedded devices, the algorithm computes a unique DRIP Score to quantify the importance of each data point. This enables dynamic decision-making on whether a data point should be stored for potential retraining or discarded without compromising model performance. Experimental evaluations on four benchmark datasets demonstrate that our approach can match or even surpass the accuracy of models trained on the entire dataset, while achieving storage savings of up to 39%. To our knowledge, this is the first algorithm to make online decisions about data point retention without requiring access to the entire dataset