A5075317929- Topic Modeling
- Online Learning and Analytics
- Natural Language Processing Techniques
- Quantum Computing Algorithms and Architecture
- Web Data Mining and Analysis
- Explainable Artificial Intelligence (XAI)
- Quantum Mechanics and Applications
- Optical measurement and interference techniques
- Image Processing Techniques and Applications
- Artificial Intelligence in Healthcare and Education
- Quantum Information and Cryptography
- Advanced Vision and Imaging
- Context-Aware Activity Recognition Systems
- Human Mobility and Location-Based Analysis
- Technology Adoption and User Behaviour
- Biomedical and Engineering Education
- Experimental Learning in Engineering
- Robotics and Automated Systems
- Service-Oriented Architecture and Web Services
- Innovative Teaching Methods
- Interactive and Immersive Displays
- Data Quality and Management
- Education and Critical Thinking Development
- Teaching and Learning Programming
- Time Series Analysis and Forecasting
German Research Centre for Artificial Intelligence
2023-2024
University of Koblenz and Landau
2024
Rheinland-Pfälzische Technische Universität Kaiserslautern-Landau
2023-2024
In the field of quantum information science and technology, representation visualization states related processes are essential for both research education. this context, a focus lies especially on ensembles few qubits. There exist many powerful representations single-qubit multiqubit systems, such as famous Bloch sphere generalizations. Here, we utilize dimensional circle notation ensembles, adapting so-called qubits idea representing <a:math...
Large language models (LLMs) have recently gained popularity. However, the impact of their general availability through ChatGPT on sensitive areas everyday life, such as education, remains unclear. Nevertheless, societal established educational methods is already being experienced by both students and educators. Our work focuses higher physics education examines problem solving strategies. In a study, with background in were assigned to solve exercises, one group having access an internet...
The increased presence of large language models (LLMs) in educational settings has ignited debates concerning negative repercussions, including overreliance and inadequate task reflection. Our work advocates moderated usage such models, designed a way that supports students encourages critical thinking. We developed two interaction methods with ChatGPT: hint-based assistance presenting multiple answer choices. In study (N=40) answering physics questions, we compared the effects our against...
In human activity recognition (HAR), the availability of substantial ground truth is necessary for training efficient models. However, acquiring pressure data through physical sensors itself can be cost-prohibitive, time-consuming. To address this critical need, we introduce Text-to-Pressure (T2P), a framework designed to generate extensive sequences from textual descriptions activities using deep learning techniques. We show that combination vector quantization sensor along with simple text...
The increased presence of large language models (LLMs) in educational settings has ignited debates concerning negative repercussions, including overreliance and inadequate task reflection. Our work advocates moderated usage such models, designed a way that supports students encourages critical thinking. We developed two interaction methods with ChatGPT: hint-based assistance presenting multiple answer choices. In study (N=40) answering physics questions, we compared the effects our against...
As distance learning becomes increasingly important and artificial intelligence tools continue to advance, automated systems for individual have attracted significant attention. However, the scarcity of open-source online that are capable providing personalized feedback has restricted widespread implementation research-based systems. In this work, we present RATsApp, an system (AFS) incorporates features such as formative feedback. The focuses on core STEM competencies mathematical...
In the rapidly evolving interdisciplinary field of quantum information science and technology, a big obstacle is necessity understanding high-level mathematics to solve complex problems. Visualizations like (dimensional) circle notation enable us visualize not only single-qubit but also multi-qubit states, entanglement, algorithms. Current findings in educational research suggest that incorporating visualizations settings problem solving can have beneficial effects on students' performance...
Medical procedures such as venipuncture and cannulation are essential for nurses require precise skills. Learning this skill, in turn, is a challenge educators due to the number of teachers per class complexity task. The study aims help students with skill acquisition alleviate educator's workload by integrating generative AI methods provide real-time feedback on medical cannulation.
In human activity recognition (HAR), the availability of substantial ground truth is necessary for training efficient models. However, acquiring pressure data through physical sensors itself can be cost-prohibitive, time-consuming. To address this critical need, we introduce Text-to-Pressure (T2P), a framework designed to generate extensive sequences from textual descriptions activities using deep learning techniques. We show that combination vector quantization sensor along with simple text...
Accurate camera calibration is crucial for various computer vision applications. However, measuring parameters in the real world challenging and arduous, there needs to be a dataset with ground truth evaluate algorithms' accuracy. In this paper, we present SynthCal, synthetic benchmarking pipeline that generates images of patterns measure enable accurate quantification algorithm performance parameter estimation. We SynthCal-generated four common patterns, two types, environments varying...
Monitoring of human activities is an essential capability many smart systems. In recent years much progress has been achieved. One the key remaining challenges availability labeled training data, in particular taking into account degree variability activities. A possible solution to leverage large scale online data repositories. This previously attempted with image and sound as both microphones cameras are widely used sensing modalities. this paper, we describe a first step towards use...
In the field of quantum information science and technology, representation visualization states related processes are essential for both research education. this context, a focus especially lies on ensembles few qubits. There exist many powerful representations single-qubit multi-qubit systems, such as famous Bloch sphere generalizations. Here, we utilize dimensional circle notation ensembles, adapting so-called qubits idea representing n-particle system in an n-dimensional space. We show...
Wearable devices have become ubiquitous in our daily lives, constantly collecting and analyzing data based on activities. However, what these measure and, therefore, the accuracy of algorithms using this can be influenced by device's placement. For instance, a smartphone handbag might not track steps as accurately one trouser pocket. The varied locations which individuals wear is underexplored. We intend to conduct cross-sectional study surveying most common ways people utilize wearables...
A still unsolved issue in human activity recognition, as it is being used many smart systems, the availability of labeled training data or, other words, information about what sensors actually are recording. possible solution to this problem build detailed semantic domain models specifying, different detail, complex compound activities. Such would allow retrieving required without necessity time-consuming labeling. On our way develop a method leverage text-based descriptions automatically...