Digital Library of Gesellschaft für Informatik e.V.
Not a member yet
40758 research outputs found
Sort by
How to Team with Your Robot? – Exploring Challenges and Opportunities for (Inclusive) Design of Human-Robot-Interaction
No full textIn the industrial sector but also in medicine and other contexts, robotic assistance is gaining importance. While use cases and users are diverse and show different characteristics, the interaction between robots and humans is often based on one primary input modality. In this workshop, we explore criteria of human-machine teaming from the perspective of human-human teaming and investigate novel approaches of multimodal interaction with robots, particularly for inclusive design. By enabling close collaboration between the workshop participants, we will (a) build a common understanding of the need for inclusive design in human-robot-interaction, (b) raise awareness of human-robot teaming aspects through an interactive experiment, (c) discuss current approaches, such as artificial intelligence or emotion recognition to, e.g. predict intentions in collaborative scenarios, and (d) elaborate on a joint workshop outcome that will, provide the human-computer interaction and human-robot collaboration communities with, e.g. recommendations for inclusive design in human-robot interaction
Communicating Uncertainty in Arrival Time Predictions for Public Transport: A Comparison of Point and Interval Forecasts
No full textIn public transport, arrival times are typically communicated as point forecasts, aiming to present precise estimates. However, current prediction models are unable to provide such precise and reliable estimates due to unpredictable events. This results in arrival times on passenger information systems appearing inaccurate due to the lack of communicated uncertainty. We therefore investigated interval forecasts as an alternative in an online study, aiming to better communicate uncertainty in arrival times. Our findings indicate that interval forecasts improve the communication of uncertainty. Further, user satisfaction was driven primarily by waiting time, and this relationship was moderated by the forecast concept. Point forecasts were only well received when the bus arrived as predicted, otherwise users preferred the broader interval forecasts. Participants valued accuracy over precision when judging arrival times
Integration des Inhaltsbereichs "Sprachen und Automaten" in die Grundschule: Gestaltung, Adaption und Auswahl "geeigneter" Aufgaben
No full textWir stellen Unterrichtsmaterial zum Inhaltsbereich "Sprachen und Automaten" für die Grundschule vor. Dieses wurde mittels Aufgaben-Analyse-Instrument entwickelt und wird hinsichtlich geeigneter Repräsentationsformen und Lernunterstützung reflektiert
Eine theoriegeleitete Analyse von Debugging-Prozessen in Scratch
No full textDebugging stellt eine große Herausforderung für Programmieranfänger:innen dar und endet nicht selten in Resignation und Frustration. In diesem Beitrag wird ein Vorgehen entwickelt, dass ermöglicht die Debugging-Prozesse von Lernenden zu untersuchen. Im Unterschied zu anderen Untersuchungen, die zumeist induktiv das Verhalten beschreiben, wird hier das Verhalten auf einen idealtypischen Prozess gemappt, um Probleme und Herausforderungen zu erkennen und Unterstützungsbedarf zu identifizieren. Dazu wurden Screenrecordings qualitativ analysiert, die die Bearbeitung von Debugging-Aufgaben in Scratch von Schüler:innen der Sekundarstufe zeigen. Aus den Ergebnissen lassen sich Indizien für einen nicht erfolgreichen Debugging-Prozess ableiten
Bildmanipulation als Lernszenario im Bereich Algorithmik
No full textIn diesem Beitrag wollen wir Möglichkeiten der Thematisierung von Bildmanipulationen als Verknüpfung der Themengebiete Algorithmik und Codierung (hier: RGB-Modell) im Informatikunterricht aufzeigen
Debugging von Physical Computing Anwendungen
No full textDieser Artikel beschreibt ein Promotionsvorhaben, welches sich mit beim Physical Computing auftretenden Fehlern auseinandergesetzt. Im Vergleich zur virtuellen Programmierung haben Fehler beim Physical Computing vielfältigere Ursachen und betreffen häufig Soft- und Hardware bzw. dem Zusammenspiel beider. In diesem Beitrag werden Fehler analysiert und systematisiert, sowohl Verknüpfungen zwischen verschiedenen Fehlerklassen erörtert. Um Fehler genauer untersuchen zu können, wird ein didaktisches Modell zur Informationsverarbeitung in Physical Computing Devices hergeleitet, auf Grundlage dessen verschiedene Interventionen abgeleitet und untersucht werden können
Embedding HCI in the Real World: Strategies for Recruitment and Field Research on Collaboration in Hybrid Spaces
No full textField studies are essential to understand the interaction between emerging technologies and users in collaborative work environments. Simultaneously, recruiting participants in this setting is particularly challenging when engaging marginalized, vulnerable, or specialized user groups. Access barriers, ethical concerns, and methodological constraints influence both the recruitment process and the overall research design. This workshop invites researchers and practitioners to share their experiences, future plans and methods from field studies and inclusive recruitment tactics on collaboration in hybrid spaces – including but not limited to healthcare settings. It also addresses methodological challenges, and fosters the collaborative development of user-centered approaches. Combining recruitment strategies with real-world field research approaches, this workshop provides a holistic perspective on conducting inclusive, methodologically robust, and ethically responsible HCI studies. Through keynotes, case studies, and participatory sessions, participants will develop strategies for effective recruitment and field study execution. By bridging these perspectives, the workshop aims to enhance methodological rigor and inclusivity in HCI research
UX to go – wie Persona Cards den Produktalltag verändern
No full textNutzerzentrierung gilt als eines der zentralen Prinzipien moderner Produktentwicklung und bildet das Herzstück des nutzerzentrierten Designprozesses (User Centered Design, UCD). In der Praxis zeigt sich jedoch häufig ein anderes Bild: Nutzer:innen werden zwar theoretisch ins Zentrum gestellt, im Projektalltag bleiben sie jedoch eine abstrakte Größe oder treten hinter technische oder organisatorische Prioritäten zurück.
Im Mittelpunkt des Beitrags steht die Frage, wie ein mittelständisches Softwareunternehmen durch die Ein-führung handlicher Persona Cards einen unter-nehmensweiten Kulturwandel hin zu stärkerem Nutzerfokus anstoßen konnte. Die Karten basieren auf Empathy Maps, wurden iterativ entwickelt und gezielt auf die spezifischen Anforderungen des firmeneigenen Produkts zugeschnitten. Im praktischen Einsatz werden die Karten im kompakten A6-Format als Fächer genutzt, sodass die zentralen Nutzergruppen jederzeit greifbar und sichtbar bleiben, als ständiger Reminder im Arbeitsalltag. Neben Design, Entwicklung und Testing fanden die Karten unerwartet auch in Sales und Business Development Anwendung.
Der Beitrag beleuchtet den Entstehungsprozess der Persona Cards, ihre vielfältigen Einsatzszenarien sowie die erzielten Wirkungen. Darauf aufbauend werden praxisnahe Empfehlungen abgeleitet, die anderen Teams helfen können, ihre Produktentwicklung stärker an den tatsächlichen Bedürfnissen der Nutzer:innen auszurichten
Inclusive External Human–Machine Interfaces (eHMIs) for Automated Vehicles: A Human-Centered Perspective
No full textAbstract—The continuous development and deployment of
autonomous vehicles (AVs) are changing the way people get
around in cities. As autonomous cars share the road with more
and more humans, a big problem arises: How can AVs let
all those who walk, cyclists, and other road users know what
they want to do, particularly those who have disabilities that
affect their vision, hearing, or mobility? Traditional vehicles
rely heavily on nonverbal signals such as eye contact, hand
gestures, and driver body language to signal intent to pedestrians.
In the absence of a human driver, AVs must assume this
communication responsibility through external human-machine
interfaces (eHMIs). However, most existing eHMI designs lean
heavily on visual or auditory output, which does not adequately
serve people with sensory or cognitive limitations.
This paper explores the transformative role of generative
artificial intelligence (AI)—such as large language models and
AI-driven media generators—and agentic AI, which is capable
of making autonomous decisions in real time based on user
and environmental context. These technologies offer promising
opportunities to create inclusive, adaptive, and intelligent eHMIs
that go beyond static signals. Instead of sending generic messages,
these systems can change how they convey them in real time
based on who is nearby, how they might interpret signals, and
the present circumstance. Our studies explore the modern-day
reputation of improvements in eHMI generation and identify
crucial regions wherein enhancements in accessibility, flexibility,
and practicality for real-world packages are necessary. We
recommend an progressive AI-pushed framework for growing
digital Human-Machine Interfaces (eHMIs) that could make use
of numerous remark modalities, which include visual, aural,
and tactile inputs, own environmental awareness, and alternate
primarily based totally on consumer interplay over time. We
additionally ensure that our method suits the dreams of the
Accessible Automated Automotive Workshop Series (A3WS),
which is an international attempt to make transportation less
complicated for every person to use. Our studies is to facilitate collaborative projects geared toward enhancing the safety,
accessibility, and fairness of self-sufficient car structures for all
users, mainly the ones who’ve been traditionally marginalized in
technological advancemen
Submission Part Replacement Strategies for Error Isolation in Java Program Evaluation
No full textAutomatic grading of Java programming exercises often relies on blackbox testing. A submitted solution may consist of several interdependent methods. An incorrect implementation of a dependent method should not automatically render the calling method incorrect. This paper addresses the challenge of error propagation in automated grading by isolating the part under test from depended-on parts using error-free replacements. Four replacement strategies – class replacement, method overriding, mocking, and bytecode manipulation – are presented and analyzed regarding their efficiency, implementation effort, security, and instructional impact. Our results show that all strategies show a usable performance and can be used in practice, however, they have characteristic advantages and disadvantages. Overall, bytecode manipulation seems to offers a favorable trade-off between performance and flexibility for both instance and static methods. These findings offer practical guidance for task authors seeking to implement effective and secure automated grading strategies