@inproceedings{59483, abstract = {{Abstract. The assessment of mechanically joined connections, such as clinched connections, is usually conducted destructively. Applicable non-destructive testing methods like computed tomography are time-consuming and costly, or, like electrical resistance measurement, provide only a limited amount of information. A fast, non-destructive evaluation of the joints condition shall be made possible by using transient dynamic analysis (TDA). It is based on the introduction of sound waves and the evaluation of the response behavior after passing through the structure. This study focuses the application of TDA to clinched shear connections to evaluate the performance of the tactile measuring setup. Twenty-one series were investigated, covering variations in joining task, manufacturing and defect. The evaluation was carried out using machine learning to determine for which series characteristic signals may be detected. It was shown that a classification of the investigated specimens is possible, whereby the classification accuracy depends on the examined variation. Furthermore, the accuracy was evaluated as a function of frequency and results were concluded to identify the limits of the used measuring setup.}}, author = {{Reschke, Gregor and Brosius, Alexander}}, booktitle = {{Materials Research Proceedings}}, issn = {{2474-395X}}, keywords = {{Joining, Machine Learning, Transient Dynamic Analysis}}, location = {{Paderborn}}, pages = {{293--300}}, publisher = {{Materials Research Forum LLC}}, title = {{{Transient dynamic analysis: Performance evaluation of tactile measurement}}}, doi = {{10.21741/9781644903551-36}}, volume = {{52}}, year = {{2025}}, } @inproceedings{62080, abstract = {{The failure behavior of fiber reinforced polymers (FRP) is strongly influenced by their microstructure, i.e. fiber arrangement or local fiber volume content. However, this information cannot be directly used for structural analyses, since it requires a discretization on micrometer level. Therefore, current failure theories do not directly account for such effects, but describe the behavior averaged over an entire specimen. This foundation in experimentally accessible loading conditions leads to purely theory based extension to more complex stress states without direct validation possibilities. This work aims at leveraging micro-scale simulations to obtain failure information under arbitrary loading conditions. The results are propagated to the meso-scale, enabling efficient structural analyses, by means of machine learning (ML). It is shown that the ML model is capable of correctly assessing previously unseen stress states and therefore poses an efficient tool of exploiting information from the micro-scale in larger simulations.}}, author = {{Gerritzen, Johannes and Hornig, Andreas and Gude, Maik}}, booktitle = {{Sheet Metal 2025}}, editor = {{Meschut, G. and Bobbert, M. and Duflou, J. and Fratini, L. and Hagenah, H. and Martins, P. and Merklein, M. and Micari, F.}}, isbn = {{978-1-64490-354-4}}, keywords = {{Failure, Fiber Reinforced Plastic, Machine Learning}}, pages = {{260–267}}, publisher = {{Materials Research Forum LLC, Materials Research Foundations}}, title = {{{Efficient failure information propagation under complex stress states in fiber reinforced polymers: From micro- to meso-scale using machine learning}}}, doi = {{10.21741/9781644903551-32}}, year = {{2025}}, } @article{58885, abstract = {{There have been several attempts to conceptualize and operationalize pedagogical content knowledge (PCK) in the context of teachers' professional competencies. A recent and popular model is the Refined Consensus Model (RCM), which proposes a framework of dispositional competencies (personal PCK—pPCK) that influence more action-related competencies (enacted PCK—ePCK) and vice versa. However, descriptions of the internal structure of pPCK and possible knowledge domains that might develop independently are still limited, being either primarily theoretically motivated or strictly hierarchical and therefore of limited use, for example, for formative feedback and further development of the RCM. Meanwhile, a non-hierarchical differentiation for the ePCK regarding the plan-teach-reflect cycle has emerged. In this study, we present an exploratory computational approach to investigate pre-service teachers' pPCK for a similar non-hierarchical structure using a large dataset of responses to a pPCK questionnaire (N=846). We drew on theoretical foundations and previous empirical findings to achieve interpretability by integrating this external knowledge into our analyses using the Computational Grounded Theory (CGT) framework. The results of a cluster analysis of the pPCK scores indicate the emergence of prototypical groups, which we refer to as competency profiles: (1) a group with low performance, (2) a group with relatively advanced competency in using pPCK to create instructional elements, (3) a group with relatively advanced competency in using pPCK to assess and analyze described instructional elements, and (4) a group with high performance. These groups show tendencies for certain language usage, which we analyze using a structural topic model in a CGT-inspired pattern refinement step. We verify these patterns by demonstrating the ability of a machine learning model to predict the competency profile assignments. Finally, we discuss some implications of the results for the further development of the RCM and their potential usability for an automated formative assessment.}}, author = {{Zeller, Jannis and Riese, Josef}}, issn = {{1098-2736}}, journal = {{Journal of Research in Science Teaching}}, keywords = {{computational grounded theory, language analysis, machine learning, pedagogical content knowledge, unsupervised learning}}, title = {{{Competency Profiles of PCK Using Unsupervised Learning: What Implications for the Structures of pPCK Emerge From Non-Hierarchical Analyses?}}}, doi = {{10.1002/tea.70001}}, year = {{2025}}, } @inproceedings{60680, abstract = {{Classical machine learning techniques often struggle with overfitting and unreliable predictions when exposed to novel conditions. Introducing causality into the modelling process offers a promising way to mitigate these challenges by enhancing predictive robustness. However, constructing an initial causal graph manually using domain knowledge is time-consuming, particularly in complex time series with numerous variables. To address this, causal discovery algorithms can provide a preliminary causal structure that domain experts can refine. This study investigates causal feature selection with domain knowledge using a data center system as an example. We use simulated time-series data to compare different causal feature selection with traditional machine-learning feature selection methods. Our results show that predictions based on causal features are more robust compared to those derived from traditional methods. These findings underscore the potential of combining causal discovery algorithms with human expertise to improve machine learning applications.}}, author = {{Zapata Gonzalez, David Ricardo and Meyer, Marcel and Müller, Oliver}}, keywords = {{Causal Machine Learning, Causality in Time Series, Causal Discovery, Human-Machine Collaboration}}, location = {{Amman, Jordan}}, title = {{{Bridging the gap between data-driven and theory-driven modelling – leveraging causal machine learning for integrative modelling of dynamical systems}}}, year = {{2025}}, } @article{55400, abstract = {{This study contributes to the evolving field of robot learning in interaction with humans, examining the impact of diverse input modalities on learning outcomes. It introduces the concept of "meta-modalities" which encapsulate additional forms of feedback beyond the traditional preference and scalar feedback mechanisms. Unlike prior research that focused on individual meta-modalities, this work evaluates their combined effect on learning outcomes. Through a study with human participants, we explore user preferences for these modalities and their impact on robot learning performance. Our findings reveal that while individual modalities are perceived differently, their combination significantly improves learning behavior and usability. This research not only provides valuable insights into the optimization of human-robot interactive task learning but also opens new avenues for enhancing the interactive freedom and scaffolding capabilities provided to users in such settings.}}, author = {{Beierling, Helen and Beierling, Robin and Vollmer, Anna-Lisa}}, journal = {{Frontiers in Robotics and AI}}, keywords = {{human-robot interaction, human-in-the-loop learning, reinforcement learning, interactive robot learning, multi-modal feedback, learning from demonstration, preference-based learning, scaffolding in robot learning}}, publisher = {{Frontiers }}, title = {{{The power of combined modalities in interactive robot learning}}}, volume = {{12}}, year = {{2025}}, } @article{61327, abstract = {{Robot learning from humans has been proposed and researched for several decades as a means to enable robots to learn new skills or adapt existing ones to new situations. Recent advances in artificial intelligence, including learning approaches like reinforcement learning and architectures like transformers and foundation models, combined with access to massive datasets, has created attractive opportunities to apply those data-hungry techniques to this problem. We argue that the focus on massive amounts of pre-collected data, and the resulting learning paradigm, where humans demonstrate and robots learn in isolation, is overshadowing a specialized area of work we term Human-Interactive-Robot-Learning (HIRL). This paradigm, wherein robots and humans interact during the learning process, is at the intersection of multiple fields (artificial intelligence, robotics, human-computer interaction, design and others) and holds unique promise. Using HIRL, robots can achieve greater sample efficiency (as humans can provide task knowledge through interaction), align with human preferences (as humans can guide the robot behavior towards their expectations), and explore more meaningfully and safely (as humans can utilize domain knowledge to guide learning and prevent catastrophic failures). This can result in robotic systems that can more quickly and easily adapt to new tasks in human environments. The objective of this paper is to provide a broad and consistent overview of HIRL research and to guide researchers toward understanding the scope of HIRL, and current open or underexplored challenges related to four themes — namely, human, robot learning, interaction, and broader context. The paper includes concrete use cases to illustrate the interaction between these challenges and inspire further research according to broad recommendations and a call for action for the growing HIRL community}}, author = {{Baraka, Kim and Idrees, Ifrah and Faulkner, Taylor Kessler and Biyik, Erdem and Booth, Serena and Chetouani, Mohamed and Grollman, Daniel H. and Saran, Akanksha and Senft, Emmanuel and Tulli, Silvia and Vollmer, Anna-Lisa and Andriella, Antonio and Beierling, Helen and Horter, Tiffany and Kober, Jens and Sheidlower, Isaac and Taylor, Matthew E. and van Waveren, Sanne and Xiao, Xuesu}}, journal = {{Transactions on Human-Robot Interaction}}, keywords = {{Robot learning, Interactive learning systems, Human-robot interaction, Human-in-the-loop machine learning, Teaching and learning}}, title = {{{Human-Interactive Robot Learning: Definition, Challenges, and Recommendations}}}, year = {{2025}}, } @inbook{62701, abstract = {{Learning continuous vector representations for knowledge graphs has significantly improved state-of-the-art performances in many challenging tasks. Yet, deep-learning-based models are only post-hoc and locally explainable. In contrast, learning Web Ontology Language (OWL) class expressions in Description Logics (DLs) is ante-hoc and globally explainable. However, state-of-the-art learners have two well-known lim-itations: scaling to large knowledge graphs and handling missing infor-mation. Here, we present a decision-tree-based learner (tDL) to learn Web Ontology Languages (OWLs) class expressions over large knowl-edge graphs, while imputing missing triples. Given positive and negative example individuals, tDL firstly constructs unique OWL expressions in .SHOIN from concise bounded descriptions of individuals. Each OWL class expression is used as a feature in a binary classification problem to represent input individuals. Thereafter, tDL fits a CART decision tree to learn Boolean decision rules distinguishing positive examples from nega-tive examples. A final OWL expression in.SHOIN is built by traversing the built CART decision tree from the root node to leaf nodes for each positive example. By this, tDL can learn OWL class expressions without exploration, i.e., the number of queries to a knowledge graph is bounded by the number of input individuals. Our empirical results show that tDL outperforms the current state-of-the-art models across datasets. Impor-tantly, our experiments over a large knowledge graph (DBpedia with 1.1 billion triples) show that tDL can effectively learn accurate OWL class expressions, while the state-of-the-art models fail to return any results. Finally, expressions learned by tDL can be seamlessly translated into natural language explanations using a pre-trained large language model and a DL verbalizer.}}, author = {{Demir, Caglar and Yekini, Moshood and Röder, Michael and Mahmood, Yasir and Ngonga Ngomo, Axel-Cyrille}}, booktitle = {{Lecture Notes in Computer Science}}, isbn = {{9783032060655}}, issn = {{0302-9743}}, keywords = {{Decision Tree, OWL Class Expression Learning, Description Logic, Knowledge Graph, Large Language Model, Verbalizer}}, location = {{Porto, Portugal}}, publisher = {{Springer Nature Switzerland}}, title = {{{Tree-Based OWL Class Expression Learner over Large Graphs}}}, doi = {{10.1007/978-3-032-06066-2_29}}, year = {{2025}}, } @inproceedings{62007, abstract = {{Ensemble methods are widely employed to improve generalization in machine learning. This has also prompted the adoption of ensemble learning for the knowledge graph embedding (KGE) models in performing link prediction. Typical approaches to this end train multiple models as part of the ensemble, and the diverse predictions are then averaged. However, this approach has some significant drawbacks. For instance, the computational overhead of training multiple models increases latency and memory overhead. In contrast, model merging approaches offer a promising alternative that does not require training multiple models. In this work, we introduce model merging, specifically weighted averaging, in KGE models. Herein, a running average of model parameters from a training epoch onward is maintained and used for predictions. To address this, we additionally propose an approach that selectively updates the running average of the ensemble model parameters only when the generalization performance improves on a validation dataset. We evaluate these two different weighted averaging approaches on link prediction tasks, comparing the state-of-the-art benchmark ensemble approach. Additionally, we evaluate the weighted averaging approach considering literal-augmented KGE models and multi-hop query answering tasks as well. The results demonstrate that the proposed weighted averaging approach consistently improves performance across diverse evaluation settings.}}, author = {{Sapkota, Rupesh and Demir, Caglar and Sharma, Arnab and Ngonga Ngomo, Axel-Cyrille}}, booktitle = {{Proceedings of the Thirteenth International Conference on Knowledge Capture(K-CAP 2025)}}, keywords = {{Knowledge Graphs, Embeddings, Ensemble Learning}}, location = {{Dayton, OH, USA}}, publisher = {{ACM}}, title = {{{Parameter Averaging in Link Prediction}}}, doi = {{https://doi.org/10.1145/3731443.3771365}}, year = {{2025}}, } @inbook{58874, author = {{Fahrbach, Manuel and Jenert, Tobias and Fust, Alexander and Bellwald, Noah and Winkler, Christoph}}, booktitle = {{Annals of Entrepreneurship Education and Pedagogy - 2025}}, isbn = {{9781035325795}}, keywords = {{Self-Regulated Learning, Entrepreneurship Education, Entrepreneurship Research}}, pages = {{249–265}}, publisher = {{Edward Elgar Publishing}}, title = {{{Fostering self-regulated entrepreneurial learning in entrepreneurship education}}}, doi = {{10.4337/9781035325795.00021}}, year = {{2025}}, } @article{63498, author = {{Kirchgässner, Wilhelm and Förster, Nikolas and Piepenbrock, Till and Schweins, Oliver and Wallscheid, Oliver}}, journal = {{IEEE Transactions on Power Electronics}}, keywords = {{Mathematical models, Estimation, Data models, Convolutional neural networks, Accuracy, Magnetic hysteresis, Magnetic cores, Temperature measurement, Magnetic domains, Temperature distribution, Convolutional neural network (CNN), machine learning (ML), magnetics}}, number = {{2}}, pages = {{3326--3335}}, title = {{{HARDCORE: H-Field and Power Loss Estimation for Arbitrary Waveforms With Residual, Dilated Convolutional Neural Networks in Ferrite Cores}}}, doi = {{10.1109/TPEL.2024.3488174}}, volume = {{40}}, year = {{2025}}, } @unpublished{53793, abstract = {{We utilize extreme learning machines for the prediction of partial differential equations (PDEs). Our method splits the state space into multiple windows that are predicted individually using a single model. Despite requiring only few data points (in some cases, our method can learn from a single full-state snapshot), it still achieves high accuracy and can predict the flow of PDEs over long time horizons. Moreover, we show how additional symmetries can be exploited to increase sample efficiency and to enforce equivariance.}}, author = {{Harder, Hans and Peitz, Sebastian}}, keywords = {{extreme learning machines, partial differential equations, data-driven prediction, high-dimensional systems}}, title = {{{Predicting PDEs Fast and Efficiently with Equivariant Extreme Learning Machines}}}, year = {{2024}}, } @inproceedings{54960, abstract = {{Das Fachdidaktische Wissen (FDW) wird als zentrale Komponente des Professionswissens von Lehrkräften bereits lange intensiv untersucht. Bislang liegen Ergebnisse zu Zusammenhängen des FDW mit anderen Professionswissensbereichen, zur Performanz in prototypischen Handlungssituationen und erste datengestützte inhaltlich-hierarchische Analysen auf Basis von Item Response Modellen (IRT-Modellen) vor. Im Zusammenhang mit einem projektübergreifend durchgeführten Vergleich entsprechender IRT-Modelle haben sich jedoch Limitationen bei der Vereinbarkeit und der inhaltlichen Reichhaltigkeit entsprechender Ergebnisse gezeigt, wie im Beitrag vorgestellt wird . Daher werden Analysemethoden aus dem Bereich des Machine Learning (unsupervised) vorgeschlagen, welche im Gegensatz zu IRT-Modellen auch nicht-hierarchische inhaltliche Strukturen aufdecken können. Es werden Ergebnisse entsprechender Clusteranalysen sowie Analysepläne zur Unterstützung dieser auf Basis der authentischen Sprachproduktionen von Proband:innen mithilfe von Natural Language Processing vorgestellt.}}, author = {{Zeller, Jannis and Riese, Josef}}, booktitle = {{Frühe naturwissenschaftliche Bildung, Tagungsband der GDCP Jahrestagung 2023}}, editor = {{van Vorst, Helena}}, keywords = {{Physikdidaktisches Wissen, Fähigkeitsprofile, Machine Learning}}, location = {{Hamburg}}, pages = {{122--125}}, publisher = {{Gesellschaft für Didaktik der Chemie und Physik}}, title = {{{Fähigkeitsprofile im Physikdidaktischen Wissen mithilfe von Machine Learning}}}, year = {{2024}}, } @unpublished{55159, abstract = {{We introduce a method based on Gaussian process regression to identify discrete variational principles from observed solutions of a field theory. The method is based on the data-based identification of a discrete Lagrangian density. It is a geometric machine learning technique in the sense that the variational structure of the true field theory is reflected in the data-driven model by design. We provide a rigorous convergence statement of the method. The proof circumvents challenges posed by the ambiguity of discrete Lagrangian densities in the inverse problem of variational calculus. Moreover, our method can be used to quantify model uncertainty in the equations of motions and any linear observable of the discrete field theory. This is illustrated on the example of the discrete wave equation and Schrödinger equation. The article constitutes an extension of our previous article arXiv:2404.19626 for the data-driven identification of (discrete) Lagrangians for variational dynamics from an ode setting to the setting of discrete pdes.}}, author = {{Offen, Christian}}, keywords = {{System identification, inverse problem of variational calculus, Gaussian process, Lagrangian learning, physics informed machine learning, geometry aware learning}}, pages = {{28}}, title = {{{Machine learning of discrete field theories with guaranteed convergence and uncertainty quantification}}}, year = {{2024}}, } @inproceedings{56948, abstract = {{Das Fachdidaktische Wissen (FDW) steht als zentrale Komponente des Professionswissens angehender Lehrkräfte bereits länger im Fokus der fachdidaktischen Forschung. Bisherige Ergebnisse zu möglichen Entwicklungsstufen oder prototypischen Ausprägungen des FDW ermöglichen eine differenzierte Einordnung von Lernenden auf Basis der Bearbeitung erprobter, validierter Testinstrumente. Diese Testinstrumente sind häufig mit offenen Antwortformaten gestaltet und die nachträgliche Schließung solcher Testinstrumente hat sich als nicht unproblematisch in Hinblick auf Validität und Authentizität erwiesen. Um ein automatisiertes reichhaltiges Assessment-System auf Basis der bisherigen Forschungsergebnisse zu entwickeln, können alternativ erprobte offene Testinstrumente in Kombination mit Machine-Learning basierten Auswertungsverfahren genutzt werden. Im Vortrag werden Ergebnisse einer entsprechenden Analyse auf Basis eines vergleichsweise großen (844 Bearbeitungen) Datensatzes präsentiert. Dabei wird ein zweistufiger Assessment Prozess, in dem zunächst die offenen Aufgaben mithilfe eines Sprachmodells bepunktet werden und anschließend aus den Bepunktungen inhaltlich reichhaltiges Feedback erstellt wird, genutzt.}}, author = {{Zeller, Jannis and Riese, Josef}}, booktitle = {{Entdecken, lehren und forschen im Schülerlabor. GDCP Jahrestagung 2024}}, keywords = {{Physikdidaktisches Wissen, Assessment, Machine Learning}}, location = {{Bochum}}, title = {{{Assessment des physikdidaktischen Wissens mithilfe von Machine Learning}}}, year = {{2024}}, } @inproceedings{62078, abstract = {{Fiber reinforced plastics (FRP) exhibit strongly non-linear deformation behavior. To capture this in simulations, intricate models with a variety of parameters are typically used. The identification of values for such parameters is highly challenging and requires in depth understanding of the model itself. Machine learning (ML) is a promising approach for alleviating this challenge by directly predicting parameters based on experimental results. So far, this works mostly for purely artificial data. In this work, two approaches to generalize to experimental data are investigated: a sequential approach, leveraging understanding of the constitutive model and a direct, purely data driven approach. This is exemplary carried out for a highly non-linear strain rate dependent constitutive model for the shear behavior of FRP.The sequential model is found to work better on both artificial and experimental data. It is capable of extracting well suited parameters from the artificial data under realistic conditions. For the experimental data, the model performance depends on the composition of the experimental curves, varying between excellently suiting and reasonable predictions. Taking the expert knowledge into account for ML-model training led to far better results than the purely data driven approach. Robustifying the model predictions on experimental data promises further improvement. }}, author = {{Gerritzen, Johannes and Hornig, Andreas and Winkler, Peter and Gude, Maik}}, booktitle = {{ECCM21 - Proceedings of the 21st European Conference on Composite Materials}}, isbn = {{978-2-912985-01-9}}, keywords = {{Direct parameter identification, Machine learning, Convolutional neural networks, Strain rate dependency, Fiber reinforced plastics, woven composites, segmentation, synthetic training data, x-ray computed tomography}}, pages = {{1252–1259}}, publisher = {{European Society for Composite Materials (ESCM)}}, title = {{{Direct parameter identification for highly nonlinear strain rate dependent constitutive models using machine learning}}}, doi = {{10.60691/yj56-np80}}, volume = {{3}}, year = {{2024}}, } @inproceedings{57895, abstract = {{In our paper, we present a study in which we investigate which strategies pre-service teachers (PSTs) use to find and, if necessary, reject possible candidates for congruence theorems for quadrilaterals. This study was conducted before the PTSs attended a university geometry course. In this way, statements about learning prerequisites can be made. For the study, we analyzed group discussions of PSTs to identify typical approaches and evaluate them from a mathematical perspective. The results can be considered for the further development of courses for PSTs and generate hypotheses for further research.}}, author = {{Hoffmann, Max and Schlüter, Sarah}}, booktitle = {{Proceedings of the Fifth Conference of the International Network for Didactic Research in University Mathematics (INDRUM 2024, 10-14 June 2024)}}, editor = {{González-Martín, Alejandro S. and Gueudet, Ghislaine and Florensa, Ignasi and Lombard, Nathan}}, keywords = {{Teachers’ and students’ practices at university level, Transition to, across and from university mathematics, Teaching and learning of specific topics in university mathematics, Congruence, Quadrilaterals}}, publisher = {{Escola Univerist`aria Salesiana de Sarri`a – Univ. Aut`onoma de Barcelona and INDRUM}}, title = {{{How Do Advanced Pre-Service Teachers Develop Congruence Theorems for Quadrilaterals?}}}, year = {{2024}}, } @inproceedings{56983, abstract = {{Detecting the veracity of a statement automatically is a challenge the world is grappling with due to the vast amount of data spread across the web. Verifying a given claim typically entails validating it within the framework of supporting evidence like a retrieved piece of text. Classifying the stance of the text with respect to the claim is called stance classification. Despite advancements in automated fact-checking, most systems still rely on a substantial quantity of labeled training data, which can be costly. In this work, we avoid the costly training or fine-tuning of models by reusing pre-trained large language models together with few-shot in-context learning. Since we do not train any model, our approach ExPrompt is lightweight, demands fewer resources than other stance classification methods and can serve as a modern baseline for future developments. At the same time, our evaluation shows that our approach is able to outperform former state-of-the-art stance classification approaches regarding accuracy by at least 2 percent. Our scripts and data used in this paper are available at https://github.com/dice-group/ExPrompt.}}, author = {{Qudus, Umair and Röder, Michael and Vollmers, Daniel and Ngonga Ngomo, Axel-Cyrille}}, booktitle = {{Proceedings of the 33rd ACM International Conference on Information and Knowledge Management}}, isbn = {{79-8-4007-0436-9/24/10}}, keywords = {{Stance Classification, Few-shot in-context learning, Pre-trained large language models}}, location = {{Boise, ID, USA}}, pages = {{3994 -- 3999}}, publisher = {{ACM}}, title = {{{ExPrompt: Augmenting Prompts Using Examples as Modern Baseline for Stance Classification}}}, doi = {{10.1145/3627673.3679923}}, volume = {{9}}, year = {{2024}}, } @inproceedings{57240, abstract = {{Validating assertions before adding them to a knowledge graph is an essential part of its creation and maintenance. Due to the sheer size of knowledge graphs, automatic fact-checking approaches have been developed. These approaches rely on reference knowledge to decide whether a given assertion is correct. Recent hybrid approaches achieve good results by including several knowledge sources. However, it is often impractical to provide a sheer quantity of textual knowledge or generate embedding models to leverage these hybrid approaches. We present FaVEL, an approach that uses algorithm selection and ensemble learning to amalgamate several existing fact-checking approaches that rely solely on a reference knowledge graph and, hence, use fewer resources than current hybrid approaches. For our evaluation, we create updated versions of two existing datasets and a new dataset dubbed FaVEL-DS. Our evaluation compares our approach to 15 fact-checking approaches—including the state-of-the-art approach HybridFC—on 3 datasets. Our results demonstrate that FaVEL outperforms all other approaches significantly by at least 0.04 in terms of the area under the ROC curve. Our source code, datasets, and evaluation results are open-source and can be found at https://github.com/dice-group/favel.}}, author = {{Qudus, Umair and Röder, Michael and Tatkeu Pekarou, Franck Lionel and Morim da Silva, Ana Alexandra and Ngonga Ngomo, Axel-Cyrille}}, booktitle = {{EKAW 2024}}, editor = {{Rospocher, Marco}}, keywords = {{fact checking, ensemble learning, transfer learning, knowledge management.}}, location = {{Amsterdam, Netherlands}}, title = {{{FaVEL: Fact Validation Ensemble Learning}}}, year = {{2024}}, } @inproceedings{61273, abstract = {{In human-machine explanation interactions, such as tutoring systems or customer support chatbots, it is important for the machine explainer to infer the human user's understanding. Nonverbal signals play an important role for expressing mental states like understanding and confusion in these interactions. However, an individual's expressions may vary depending on other factors. In cases where these factors are unknown, machine learning methods that infer understanding from nonverbal cues become unreliable. Stress for example has been shown to affect human expression, but it is not clear from the current research how stress affects the expression of understanding. To address this gap, we design a paradigm that induces understanding and confusion through game rule explanations. During the explanations, self-perceived understanding and confusion are annotated by the participants. A stress condition is also introduced to enable the investigation of changes in the expression of social signals under stress. We conducted a study to validate the stress induction and participants reported a statistically significant increase in stress during the stress condition compared to the neutral control condition. Additionally, feedback from participants shows that the paradigm is effective in inducing understanding and confusion. This paradigm paves the way for further studies investigating social signals of understanding to improve human-machine explanation interactions for varying contexts.}}, author = {{Paletschek, Jonas}}, booktitle = {{12th International Conference on Affective Computing & Intelligent Interaction}}, keywords = {{Understanding, Nonverbal Social Signals, Stress Induction, Explanation, Machine Learning Bias}}, location = {{Glasgow}}, publisher = {{IEEE}}, title = {{{A Paradigm to Investigate Social Signals of Understanding and Their Susceptibility to Stress}}}, doi = {{10.1109/ACII63134.2024.00040}}, year = {{2024}}, } @article{55999, abstract = {{Clean hydrogen is a key aspect of carbon neutrality, necessitating robust methods for monitoring hydrogen concentration in critical infrastructures like pipelines or power plants. While semiconducting metal oxides such as In2O3 can monitor gas concentrations down to the ppm range, they often exhibit cross-sensitivity to other gases like H2O. In this study, we investigated whether cyclic optical illumination of a gas-sensitive In2O3 layer creates identifiable changes in a gas sensor´s electronic resistance that can be linked to H2 and H2O concentrations via machine learning. We exposed nanostructured In2O3 with a large surface area of 95 m2 g-1 to H2 concentrations (0-800 ppm) and relative humidity (0-70%) under cyclic activation utilizing blue light. The sensors were tested for 20 classes of gas combinations. A support vector machine achieved classification rates up to 92.0%, with reliable reproducibility (88.2 ± 2.7%) across five individual sensors using 10-fold cross-validation. Our findings suggest that cyclic optical activation can be used as a tool to classify H2 and H2O concentrations.}}, author = {{Baier, Dominik and Krüger, Alexander and Wagner, Thorsten and Tiemann, Michael and Weinberger, Christian}}, issn = {{2227-9040}}, journal = {{Chemosensors}}, keywords = {{resistive gas sensor, chemiresistor, semiconductor, metal oxide, In2O3, mesoporous, hydrogen, humidtiy, machine learning, sustainable}}, number = {{9}}, pages = {{178}}, publisher = {{MDPI}}, title = {{{Gas Sensing with Nanoporous In2O3 under Cyclic Optical Activation: Machine Learning-Aided Classification of H2 and H2O}}}, doi = {{10.3390/chemosensors12090178}}, volume = {{12}}, year = {{2024}}, }