Refine
Year of publication
- 2023 (43) (remove)
Document Type
- Doctoral Thesis (17)
- Part of Periodical (13)
- Master's Thesis (7)
- Article (4)
- Other (2)
Keywords
- Action Recognition (1)
- Action Segmentation (1)
- Anforderung (1)
- Antagonistic interactions (1)
- Axiomatische Geometrie (1)
- Barbatula barbatula (1)
- Bauxit (1)
- Berufsorientierung (1)
- CLIL (1)
- Computer assisted communication (1)
- Cottus gobio (1)
- Curriculum (1)
- Design Science Research (1)
- Eigenschaften von Teammitgliedern (1)
- Elternschaft (1)
- Employee Behavior (1)
- Erziehungswissenschaft (1)
- Euklidische Geometrie (1)
- Faxonius limosus (1)
- Fernarbeitsregelungen (1)
- Flesch-Reading-Ease Index (1)
- Forschen in Koblenz (1)
- Fremdsprachendidaktik (1)
- Fächerkanon (1)
- Gamification (1)
- Geometriedidaktik (1)
- Grayscale (1)
- Hochschuldidaktik (1)
- Homeoffice (1)
- Imitation Learning (1)
- Junges Forschen (1)
- KSAO (1)
- Klimawandel (1)
- Kooperation (1)
- Kulturelle Bildung (1)
- Laugung (1)
- Leichte Sprache (1)
- Linguistic Requirements (1)
- Ländliche Räume (1)
- Mobile Arbeit (1)
- Multimodal Action Recognition (1)
- One-Shot Action Recognition (1)
- Praktikant (1)
- Praktikum (1)
- Praktikumsbegleitung (1)
- Problematic smartphone use (1)
- Projektpraktikum (1)
- Pädagogik (1)
- Regionalbanken (1)
- Smartphone addiction (1)
- Spiralcurriculum (1)
- Statistische Versuchsplanung (1)
- Studiengang (1)
- Studium (1)
- Suffering (1)
- Sustainability (1)
- Telearbeit (1)
- Types of smartphone use (1)
- Virtuelle Teams (1)
- Virtuelle Zusammenarbeit (1)
- Wagenschein (1)
- Wissenschaft Koblenz (1)
- X-ray computer tomography (XRT) (1)
- Zusammenarbeit (1)
- acid leaching (1)
- age cohorts (1)
- bauxite (1)
- empirische Untersuchung (1)
- excess deaths (1)
- invasive crayfish (1)
- iron removal (1)
- lexical sophistication (1)
- nachhaltigkeitsorientierter Führungsstil (1)
- natural language generation (1)
- plain language (1)
- prognosis model (1)
- refractory grade (1)
- scaffolded writing (1)
Institute
- Zentrale Einrichtungen (11)
- Institut für Management (9)
- Institut für Computervisualistik (3)
- Institut für Integrierte Naturwissenschaften, Abt. Chemie (3)
- Mathematisches Institut (3)
- Institut für Informatik (2)
- Institut für Pädagogik, Abteilung Pädagogik (2)
- Institut für Anglistik und Amerikanistik (1)
- Institut für Integrierte Naturwissenschaften (1)
- Institut für Integrierte Naturwissenschaften, Abt. Biologie (1)
Artificial neural networks is a popular field of research in artificial intelli-
gence. The increasing size and complexity of huge models entail certain
problems. The lack of transparency of the inner workings of a neural net-
work makes it difficult to choose efficient architectures for different tasks.
It proves to be challenging to solve these problems, and with a lack of in-
sightful representations of neural networks, this state of affairs becomes
entrenched. With these difficulties in mind a novel 3D visualization tech-
nique is introduced. Attributes for trained neural networks are estimated
by utilizing established methods from the area of neural network optimiza-
tion. Batch normalization is used with fine-tuning and feature extraction to
estimate the importance of different parts of the neural network. A combi-
nation of the importance values with various methods like edge bundling,
ray tracing, 3D impostor and a special transparency technique results in a
3D model representing a neural network. The validity of the extracted im-
portance estimations is demonstrated and the potential of the developed
visualization is explored.
Counts of SARS-CoV-2-related deaths have been key numbers for justifying severe political, social and economical measures imposed by authorities world-wide. A particular focus thereby was the concomitant excess mortality (EM), i.e. fatalities above the expected all-cause mortality (AM). Recent studies, inter alia by the WHO, estimated the SARS-CoV-2-related EM in Germany between 2020 and 2021 as high as 200 000. In this study, we attempt to scrutinize these numbers by putting them into the context of German AM since the year 2000. We propose two straightforward, age-cohort-dependent models to estimate German AM for the ‘Corona pandemic’ years, as well as the corresponding flu seasons, out of historic data. For Germany, we find overall negative EM of about −18 500 persons for the year 2020, and a minor positive EM of about 7000 for 2021, unveiling that officially reported EM counts are an exaggeration. In 2022, the EM count is about 41 200. Further, based on NAA-test-positive related death counts, we are able to estimate how many Germans have died due to rather than with CoViD-19; an analysis not provided by the appropriate authority, the RKI. Through 2020 and 2021 combined, our due estimate is at no more than 59 500. Varying NAA test strategies heavily obscured SARS-CoV-2-related EM, particularly within the second year of the proclaimed pandemic. We compensated changes in test strategies by assuming that age-cohort-specific NAA-conditional mortality rates during the first pandemic year reflected SARS-CoV-2-characteristic constants.
X-ray computer tomography (XRT) is a three-dimensional, nondestructive, and thus reproducible examination method that allows for the investigation of internal and external structures of objects. Due to its characteristics, the XRT technique has increasingly established itself as an alternative examination method and is also applied in the field of mineral processing. Within this work, XRT is used to investigate the influence of hydrochloric acid leaching of iron-rich bauxites on grain composition. Acid leaching is a promising method for the beneficiation of iron-rich bauxites for refractories. Many studies have already established that leaching with hydrochloric acid can reduce the Fe₂O₃ content in bauxites. However, apart from the influence of the leaching process on the composition of the bauxites, aspects such as the influence of the acid on the exact grain constitution or the porosity behavior have rarely been considered so far. To address these open questions, XRT analysis was used to examine and characterize various bauxites. By comparing identical grains before and after leaching, it was observed that in gibbsite bauxites the acid penetration is deeper, and the volume decreases significantly. In diasporic and boehmitic bauxites, clear leaching edges can be seen in which the iron content has been reduced.
Die Umsetzung von Fernarbeitsregelungen wurde bisher überwiegend in allgemeiner Natur betrachtet. In der vorliegenden Arbeit wird mit Hilfe von Interviews mit Führungskräften, Mitarbeitern und IT-Dienstleistern die Umsetzung von Fernarbeitsregelungen in deutschen Regionalbanken betrachtet. Mit Hilfe der Grounded Theory werden dabei Herausforderungen, Lösungsansätze, Vorteile und Nachteile einer Umsetzung herausgearbeitet. Dabei zeigt sich, dass die Umsetzung zwischen den verschiedenen Banken starke Unterschiede aufweist, was den Fortschritt betrifft. Besonders die Überzeugung, dass eine Umsetzung vorteilhaft für die Bank sowie die Mitarbeiter ist, zeigt sich als Kernproblem in der eher konservativen Bankenlandschaft. Mit Hilfe der Arbeit werden Lösungsansätze gezeigt, um Herausforderungen bewältigen zu können und den Einfluss von Nachteilen zu minimieren. Dadurch dient die Arbeit dazu, Regionalbanken bei einer Umsetzung von Fernarbeitsregelungen zu unterstützen, indem sie Potenziale erkennen können und Möglichkeiten aufgezeigt bekommen, eine Umsetzung zu gewährleisten.
Digital transformation is a prevailing trend in the world, especially in dynamic Asia. Vietnam has recorded remarkable changes in the economy as domestic enterprises have made new strides in the digital transformation process. MB Bank, one of the prestigious financial groups in Vietnam, also takes advantage of digital transformation to have the opportunity to break through to become a large-scale technology enterprise with many factors such as improving customer experience, increasing customer base and increasing customer satisfaction. enhance competitiveness, build trust and loyalty for customers. However, in the process of converting MB, there are also many challenges that require banks to have appropriate policies to handle. It can be said that MB Bank is a typical case study of digital transformation in the banking sector in Vietnam.
Challenges of Implementing Innovation Strategies at Large Organizations: A case of Lotte Group
(2023)
For many decades, one of the most important focuses of research has been on determining whether or not there is a correlation between the size of an organization and its level of innovation. Unlike small companies, large companies often have well-established structure that are hard to change and change managements seems to be much more difficult especially related to innovation. Nevertheless, there are many examples to prove the opposites. Some large organization like Apple, Amazon... always show great innovation efforts and keep changing in a much positive way. Therefore, the aim of this thesis is to discuss of how large organization can be able to implement innovation when having much drawbacks compare to SMEs. Through the use of a qualitative research approach, researcher was able to explore essential information on the innovation strategies that large companies are using in order to innovate and how they could overcome existing challenges by studying the working process of Lotte Group – one of the biggest companies in Korea.
FinTech is deemed to be an underexplored phenomenon even in academic and real environments. Among (1) “Sustainable FinTech” – the application of information technology as innovation in established financial services providers’ business operation; and (2) “Disruptive FinTech” – the provision of financial products and services by non-incumbents which in most cases are information technology entrepreneurs, the former receives more attention. In order to contribute to Disruptive FinTech category, the thesis strive to examine Entrepreneurial Strategy framework applied for technology players taking part in Vietnam financial market.
Potential impacts of invasive crayfish on native
benthic fish: shelter use and agonistic behaviour
(2023)
Spinycheek crayfish (Faxonius limosus) and signal crayfish (Pacifastacus leniusculus) are successful North American invasive crayfish species distributed throughout Europe. Both species compete with native benthic fish for shelter. In a laboratory approach, we assessed competition for shelter and antagonistic interactions between these invasive crayfish species and the native benthic fish species, stone loach (Barbatula barbatula) and bullhead (Cottus gobio). This allows for studying the potential impacts of invasive crayfish on native benthic fish. Spinycheek crayfish and signal crayfish were able to gain control of the shelter and could successfully displace both benthic fish species. For stone loach, the presence of crayfish significantly decreased their shelter use and caused several behavioural changes such as reduced activity and increased hiding behaviour outside the shelter. Although the shelter use by bullheads was not reduced, they displayed similar behavioural changes, if less intense. Invasive crayfish species showed remarkable combative interactions against both species of benthic fishes, evidenced by the high number of aggressive interactions, especially concerning stone loach. Our results highlight the pronounced dominance of invasive crayfish over benthic fish in terms of shelter competition and aggressive interactions under laboratory conditions, which consequently might promote the latter’s exposure to predation.
On the recognition of human activities and the evaluation of its imitation by robotic systems
(2023)
This thesis addresses the problem of action recognition through the analysis of human motion and the benchmarking of its imitation by robotic systems.
For our action recognition related approaches, we focus on presenting approaches that generalize well across different sensor modalities. We transform multivariate signal streams from various sensors to a common image representation. The action recognition problem on sequential multivariate signal streams can then be reduced to an image classification task for which we utilize recent advances in machine learning. We demonstrate the broad applicability of our approaches formulated as a supervised classification task for action recognition, a semi-supervised classification task for one-shot action recognition, modality fusion and temporal action segmentation.
For action classification, we use an EfficientNet Convolutional Neural Network (CNN) model to classify the image representations of various data modalities. Further, we present approaches for filtering and the fusion of various modalities on a representation level. We extend the approach to be applicable for semi-supervised classification and train a metric-learning model that encodes action similarity. During training, the encoder optimizes the distances in embedding space for self-, positive- and negative-pair similarities. The resulting encoder allows estimating action similarity by calculating distances in embedding space. At training time, no action classes from the test set are used.
Graph Convolutional Network (GCN) generalized the concept of CNNs to non-Euclidean data structures and showed great success for action recognition directly operating on spatio-temporal sequences like skeleton sequences. GCNs have recently shown state-of-the-art performance for skeleton-based action recognition but are currently widely neglected as the foundation for the fusion of various sensor modalities. We propose incorporating additional modalities, like inertial measurements or RGB features, into a skeleton-graph, by proposing fusion on two different dimensionality levels. On a channel dimension, modalities are fused by introducing additional node attributes. On a spatial dimension, additional nodes are incorporated into the skeleton-graph.
Transformer models showed excellent performance in the analysis of sequential data. We formulate the temporal action segmentation task as an object detection task and use a detection transformer model on our proposed motion image representations. Experiments for our action recognition related approaches are executed on large-scale publicly available datasets. Our approaches for action recognition for various modalities, action recognition by fusion of various modalities, and one-shot action recognition demonstrate state-of-the-art results on some datasets.
Finally, we present a hybrid imitation learning benchmark. The benchmark consists of a dataset, metrics, and a simulator integration. The dataset contains RGB-D image sequences of humans performing movements and executing manipulation tasks, as well as the corresponding ground truth. The RGB-D camera is calibrated against a motion-capturing system, and the resulting sequences serve as input for imitation learning approaches. The resulting policy is then executed in the simulated environment on different robots. We propose two metrics to assess the quality of the imitation. The trajectory metric gives insights into how close the execution was to the demonstration. The effect metric describes how close the final state was reached according to the demonstration. The Simitate benchmark can improve the comparability of imitation learning approaches.