Springer
Refine
Year of publication
Institute
- Fachbereich Elektrotechnik und Informationstechnik (80)
- Fachbereich Medizintechnik und Technomathematik (79)
- IfB - Institut für Bioengineering (44)
- Fachbereich Luft- und Raumfahrttechnik (35)
- Fachbereich Chemie und Biotechnologie (25)
- Fachbereich Energietechnik (25)
- Fachbereich Wirtschaftswissenschaften (21)
- Fachbereich Maschinenbau und Mechatronik (15)
- INB - Institut für Nano- und Biotechnologien (12)
- MASKOR Institut für Mobile Autonome Systeme und Kognitive Robotik (12)
Document Type
- Article (151)
- Part of a Book (104)
- Conference Proceeding (30)
- Book (4)
Keywords
- Natural language processing (4)
- MINLP (3)
- Optimization (3)
- Prozessautomatisierung (3)
- Seismic design (3)
- Additive manufacturing (2)
- CFD (2)
- Digitale Transformation (2)
- Engineering optimization (2)
- Information extraction (2)
Easy-read and large language models: on the ethical dimensions of LLM-based text simplification
(2024)
The production of easy-read and plain language is a challenging task, requiring well-educated experts to write context-dependent simplifications of texts. Therefore, the domain of easy-read and plain language is currently restricted to the bare minimum of necessary information. Thus, even though there is a tendency to broaden the domain of easy-read and plain language, the inaccessibility of a significant amount of textual information excludes the target audience from partaking or entertainment and restricts their ability to live life autonomously. Large language models can solve a vast variety of natural language tasks, including the simplification of standard language texts to easy-read or plain language. Moreover, with the rise of generative models like GPT, easy-read and plain language may be applicable to all kinds of natural language texts, making formerly inaccessible information accessible to marginalized groups like, a.o., non-native speakers, and people with mental disabilities. In this paper, we argue for the feasibility of text simplification and generation in that context, outline the ethical dimensions, and discuss the implications for researchers in the field of ethics and computer science.
The quest for scientifically advanced and sustainable solutions is driven by growing environmental and economic issues associated with coal mining, processing, and utilization. Consequently, within the coal industry, there is a growing recognition of the potential of microbial applications in fostering innovative technologies. Microbial-based coal solubilization, coal beneficiation, and coal dust suppression are green alternatives to traditional thermochemical and leaching technologies and better meet the need for ecologically sound and economically viable choices. Surfactant-mediated approaches have emerged as powerful tools for modeling, simulation, and optimization of coal-microbial systems and continue to gain prominence in clean coal fuel production, particularly in microbiological co-processing, conversion, and beneficiation. Surfactants (surface-active agents) are amphiphilic compounds that can reduce surface tension and enhance the solubility of hydrophobic molecules. A wide range of surfactant properties can be achieved by either directly influencing microbial growth factors, stimulants, and substrates or indirectly serving as frothers, collectors, and modifiers in the processing and utilization of coal. This review highlights the significant biotechnological potential of surfactants by providing a thorough overview of their involvement in coal biodegradation, bioprocessing, and biobeneficiation, acknowledging their importance as crucial steps in coal consumption.
Dieses Buch lädt dazu ein, die Welt um uns herum aus einem neuen Blickwinkel zu betrachten und dabei die spannende Verbindung zwischen der Mathematik und unserem täglichen Leben zu entdecken – denn um die Technologien und Entwicklungen unserer modernen Gesellschaft zu verstehen, benötigen wir ein intuitives Verständnis grundlegender mathematischer Ideen. In diesem Buch geht es um diese Grundlagen, vor allem aber um ihre praktische Anwendung im Alltag: Gemeinsam begeben wir uns auf eine unterhaltsame Reise und entdecken dabei, wie Mathematik in vielfältiger Weise allgegenwärtig ist. Anschauliche Beispiele zeigen, wie wir täglich – oft unbewusst – mathematische Ideen nutzen und wie wir mit Hilfe von Mathematik bessere Entscheidungen treffen können.
Nach einer Einführung in Algorithmen und Optimierungsprobleme, geht es im weiteren Verlauf um die Modellierung von Zufall und Unsicherheiten. Zum Ende des Buchs werden die Themen zusammengeführt und Algorithmen für Anwendungen besprochen, bei denen der Zufall eine entscheidende Rolle spielt.
The book covers various numerical field simulation methods, nonlinear circuit technology and its MF-S- and X-parameters, as well as state-of-the-art power amplifier techniques. It also describes newly presented oscillators and the emerging field of GHz plasma technology. Furthermore, it addresses aspects such as waveguides, mixers, phase-locked loops, antennas, and propagation effects, in combination with the bachelor's book 'High-Frequency Engineering,' encompassing all aspects related to the current state of GHz technology.
We conducted a scoping review for active learning in the domain of natural language processing (NLP), which we summarize in accordance with the PRISMA-ScR guidelines as follows:
Objective: Identify active learning strategies that were proposed for entity recognition and their evaluation environments (datasets, metrics, hardware, execution time).
Design: We used Scopus and ACM as our search engines. We compared the results with two literature surveys to assess the search quality. We included peer-reviewed English publications introducing or comparing active learning strategies for entity recognition.
Results: We analyzed 62 relevant papers and identified 106 active learning strategies. We grouped them into three categories: exploitation-based (60x), exploration-based (14x), and hybrid strategies (32x). We found that all studies used the F1-score as an evaluation metric. Information about hardware (6x) and execution time (13x) was only occasionally included. The 62 papers used 57 different datasets to evaluate their respective strategies. Most datasets contained newspaper articles or biomedical/medical data. Our analysis revealed that 26 out of 57 datasets are publicly accessible.
Conclusion: Numerous active learning strategies have been identified, along with significant open questions that still need to be addressed. Researchers and practitioners face difficulties when making data-driven decisions about which active learning strategy to adopt. Conducting comprehensive empirical comparisons using the evaluation environment proposed in this study could help establish best practices in the domain.
Mathematical morphology is a part of image processing that has proven to be fruitful for numerous applications. Two main operations in mathematical morphology are dilation and erosion. These are based on the construction of a supremum or infimum with respect to an order over the tonal range in a certain section of the image. The tonal ordering can easily be realised in grey-scale morphology, and some morphological methods have been proposed for colour morphology. However, all of these have certain limitations.
In this paper we present a novel approach to colour morphology extending upon previous work in the field based on the Loewner order. We propose to consider an approximation of the supremum by means of a log-sum exponentiation introduced by Maslov. We apply this to the embedding of an RGB image in a field of symmetric 2x2 matrices. In this way we obtain nearly isotropic matrices representing colours and the structural advantage of transitivity. In numerical experiments we highlight some remarkable properties of the proposed approach.
Fields of asymmetric tensors play an important role in many applications such as medical imaging (diffusion tensor magnetic resonance imaging), physics, and civil engineering (for example Cauchy-Green-deformation tensor, strain tensor with local rotations, etc.). However, such asymmetric tensors are usually symmetrized and then further processed. Using this procedure results in a loss of information. A new method for the processing of asymmetric tensor fields is proposed restricting our attention to tensors of second-order given by a 2x2 array or matrix with real entries. This is achieved by a transformation resulting in Hermitian matrices that have an eigendecomposition similar to symmetric matrices. With this new idea numerical results for real-world data arising from a deformation of an object by external forces are given. It is shown that the asymmetric part indeed contains valuable information.
The hot spots conjecture is only known to be true for special geometries. This paper shows numerically that the hot spots conjecture can fail to be true for easy to construct bounded domains with one hole. The underlying eigenvalue problem for the Laplace equation with Neumann boundary condition is solved with boundary integral equations yielding a non-linear eigenvalue problem. Its discretization via the boundary element collocation method in combination with the algorithm by Beyn yields highly accurate results both for the first non-zero eigenvalue and its corresponding eigenfunction which is due to superconvergence. Additionally, it can be shown numerically that the ratio between the maximal/minimal value inside the domain and its maximal/minimal value on the boundary can be larger than 1 + 10− 3. Finally, numerical examples for easy to construct domains with up to five holes are provided which fail the hot spots conjecture as well.
Generating synthetic LiDAR point cloud data for object detection using the Unreal Game Engine
(2024)
Object detection based on artificial intelligence is ubiquitous in today’s computer vision research and application. The training of the neural networks for object detection requires large and high-quality datasets. Besides datasets based on image data, datasets derived from point clouds offer several advantages. However, training datasets are sparse and their generation requires a lot of effort, especially in industrial domains. A solution to this issue offers the generation of synthetic point cloud data. Based on the design science research method, the work at hand proposes an approach and its instantiation for generating synthetic point cloud data based on the Unreal Engine. The point cloud quality is evaluated by comparing the synthetic cloud to a real-world point cloud. Within a practical example the applicability of the Unreal Game engine for synthetic point cloud generation could be successfully demonstrated.
