TY - JOUR A1 - Garbulowski, Mateusz A1 - Smolinska, Karolina A1 - Çabuk, Uğur A1 - Yones, Sara A. A1 - Celli, Ludovica A1 - Yaz, Esma Nur A1 - Barrenas, Fredrik A1 - Diamanti, Klev A1 - Wadelius, Claes A1 - Komorowski, Jan T1 - Machine learning-based analysis of glioma grades reveals co-enrichment JF - Cancers N2 - Simple Summary Gliomas are heterogenous types of cancer, therefore the therapy should be personalized and targeted toward specific pathways. We developed a methodology that corrected strong batch effects from The Cancer Genome Atlas datasets and estimated glioma grade-specific co-enrichment mechanisms using machine learning. Our findings created hypotheses for annotations, e.g., pathways, that should be considered as therapeutic targets. Gliomas develop and grow in the brain and central nervous system. Examining glioma grading processes is valuable for improving therapeutic challenges. One of the most extensive repositories storing transcriptomics data for gliomas is The Cancer Genome Atlas (TCGA). However, such big cohorts should be processed with caution and evaluated thoroughly as they can contain batch and other effects. Furthermore, biological mechanisms of cancer contain interactions among biomarkers. Thus, we applied an interpretable machine learning approach to discover such relationships. This type of transparent learning provides not only good predictability, but also reveals co-predictive mechanisms among features. In this study, we corrected the strong and confounded batch effect in the TCGA glioma data. We further used the corrected datasets to perform comprehensive machine learning analysis applied on single-sample gene set enrichment scores using collections from the Molecular Signature Database. Furthermore, using rule-based classifiers, we displayed networks of co-enrichment related to glioma grades. Moreover, we validated our results using the external glioma cohorts. We believe that utilizing corrected glioma cohorts from TCGA may improve the application and validation of any future studies. Finally, the co-enrichment and survival analysis provided detailed explanations for glioma progression and consequently, it should support the targeted treatment. KW - glioma KW - machine learning KW - batch effect KW - TCGA KW - co-enrichment KW - rough sets Y1 - 2022 U6 - https://doi.org/10.3390/cancers14041014 SN - 2072-6694 VL - 14 IS - 4 PB - MDPI CY - Basel ER - TY - THES A1 - Taleb, Aiham T1 - Self-supervised deep learning methods for medical image analysis T1 - Selbstüberwachte Deep Learning Methoden für die medizinische Bildanalyse N2 - Deep learning has seen widespread application in many domains, mainly for its ability to learn data representations from raw input data. Nevertheless, its success has so far been coupled with the availability of large annotated (labelled) datasets. This is a requirement that is difficult to fulfil in several domains, such as in medical imaging. Annotation costs form a barrier in extending deep learning to clinically-relevant use cases. The labels associated with medical images are scarce, since the generation of expert annotations of multimodal patient data at scale is non-trivial, expensive, and time-consuming. This substantiates the need for algorithms that learn from the increasing amounts of unlabeled data. Self-supervised representation learning algorithms offer a pertinent solution, as they allow solving real-world (downstream) deep learning tasks with fewer annotations. Self-supervised approaches leverage unlabeled samples to acquire generic features about different concepts, enabling annotation-efficient downstream task solving subsequently. Nevertheless, medical images present multiple unique and inherent challenges for existing self-supervised learning approaches, which we seek to address in this thesis: (i) medical images are multimodal, and their multiple modalities are heterogeneous in nature and imbalanced in quantities, e.g. MRI and CT; (ii) medical scans are multi-dimensional, often in 3D instead of 2D; (iii) disease patterns in medical scans are numerous and their incidence exhibits a long-tail distribution, so it is oftentimes essential to fuse knowledge from different data modalities, e.g. genomics or clinical data, to capture disease traits more comprehensively; (iv) Medical scans usually exhibit more uniform color density distributions, e.g. in dental X-Rays, than natural images. Our proposed self-supervised methods meet these challenges, besides significantly reducing the amounts of required annotations. We evaluate our self-supervised methods on a wide array of medical imaging applications and tasks. Our experimental results demonstrate the obtained gains in both annotation-efficiency and performance; our proposed methods outperform many approaches from related literature. Additionally, in case of fusion with genetic modalities, our methods also allow for cross-modal interpretability. In this thesis, not only we show that self-supervised learning is capable of mitigating manual annotation costs, but also our proposed solutions demonstrate how to better utilize it in the medical imaging domain. Progress in self-supervised learning has the potential to extend deep learning algorithms application to clinical scenarios. N2 - Deep Learning findet in vielen Bereichen breite Anwendung, vor allem wegen seiner Fähigkeit, Datenrepräsentationen aus rohen Eingabedaten zu lernen. Dennoch war der Erfolg bisher an die Verfügbarkeit großer annotatierter Datensätze geknüpft. Dies ist eine Anforderung, die in verschiedenen Bereichen, z. B. in der medizinischen Bildgebung, schwer zu erfüllen ist. Die Kosten für die Annotation stellen ein Hindernis für die Ausweitung des Deep Learning auf klinisch relevante Anwendungsfälle dar. Die mit medizinischen Bildern verbundenen Annotationen sind rar, da die Erstellung von Experten Annotationen für multimodale Patientendaten in großem Umfang nicht trivial, teuer und zeitaufwändig ist. Dies unterstreicht den Bedarf an Algorithmen, die aus den wachsenden Mengen an unbeschrifteten Daten lernen. Selbstüberwachte Algorithmen für das Repräsentationslernen bieten eine mögliche Lösung, da sie die Lösung realer (nachgelagerter) Deep-Learning-Aufgaben mit weniger Annotationen ermöglichen. Selbstüberwachte Ansätze nutzen unannotierte Stichproben, um generisches Eigenschaften über verschiedene Konzepte zu erlangen und ermöglichen so eine annotationseffiziente Lösung nachgelagerter Aufgaben. Medizinische Bilder stellen mehrere einzigartige und inhärente Herausforderungen für existierende selbstüberwachte Lernansätze dar, die wir in dieser Arbeit angehen wollen: (i) medizinische Bilder sind multimodal, und ihre verschiedenen Modalitäten sind von Natur aus heterogen und in ihren Mengen unausgewogen, z.B. (ii) medizinische Scans sind mehrdimensional, oft in 3D statt in 2D; (iii) Krankheitsmuster in medizinischen Scans sind zahlreich und ihre Häufigkeit weist eine Long-Tail-Verteilung auf, so dass es oft unerlässlich ist, Wissen aus verschiedenen Datenmodalitäten, z. B. Genomik oder klinische Daten, zu verschmelzen, um Krankheitsmerkmale umfassender zu erfassen; (iv) medizinische Scans weisen in der Regel eine gleichmäßigere Farbdichteverteilung auf, z. B. in zahnmedizinischen Röntgenaufnahmen, als natürliche Bilder. Die von uns vorgeschlagenen selbstüberwachten Methoden adressieren diese Herausforderungen und reduzieren zudem die Menge der erforderlichen Annotationen erheblich. Wir evaluieren unsere selbstüberwachten Methoden in verschiedenen Anwendungen und Aufgaben der medizinischen Bildgebung. Unsere experimentellen Ergebnisse zeigen, dass die von uns vorgeschlagenen Methoden sowohl die Effizienz der Annotation als auch die Leistung steigern und viele Ansätze aus der verwandten Literatur übertreffen. Darüber hinaus ermöglichen unsere Methoden im Falle der Fusion mit genetischen Modalitäten auch eine modalübergreifende Interpretierbarkeit. In dieser Arbeit zeigen wir nicht nur, dass selbstüberwachtes Lernen in der Lage ist, die Kosten für manuelle Annotationen zu senken, sondern auch, wie man es in der medizinischen Bildgebung besser nutzen kann. Fortschritte beim selbstüberwachten Lernen haben das Potenzial, die Anwendung von Deep-Learning-Algorithmen auf klinische Szenarien auszuweiten. KW - Artificial Intelligence KW - machine learning KW - unsupervised learning KW - representation learning KW - Künstliche Intelligenz KW - maschinelles Lernen KW - Representationlernen KW - selbstüberwachtes Lernen Y1 - 2024 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:517-opus4-644089 ER - TY - JOUR A1 - Shams, Boshra A1 - Wang, Ziqian A1 - Roine, Timo A1 - Aydogan, Dogu Baran A1 - Vajkoczy, Peter A1 - Lippert, Christoph A1 - Picht, Thomas A1 - Fekonja, Lucius Samo T1 - Machine learning-based prediction of motor status in glioma patients using diffusion MRI metrics along the corticospinal tract JF - Brain communications N2 - Shams et al. report that glioma patients' motor status is predicted accurately by diffusion MRI metrics along the corticospinal tract based on support vector machine method, reaching an overall accuracy of 77%. They show that these metrics are more effective than demographic and clinical variables. Along tract statistics enables white matter characterization using various diffusion MRI metrics. These diffusion models reveal detailed insights into white matter microstructural changes with development, pathology and function. Here, we aim at assessing the clinical utility of diffusion MRI metrics along the corticospinal tract, investigating whether motor glioma patients can be classified with respect to their motor status. We retrospectively included 116 brain tumour patients suffering from either left or right supratentorial, unilateral World Health Organization Grades II, III and IV gliomas with a mean age of 53.51 +/- 16.32 years. Around 37% of patients presented with preoperative motor function deficits according to the Medical Research Council scale. At group level comparison, the highest non-overlapping diffusion MRI differences were detected in the superior portion of the tracts' profiles. Fractional anisotropy and fibre density decrease, apparent diffusion coefficient axial diffusivity and radial diffusivity increase. To predict motor deficits, we developed a method based on a support vector machine using histogram-based features of diffusion MRI tract profiles (e.g. mean, standard deviation, kurtosis and skewness), following a recursive feature elimination method. Our model achieved high performance (74% sensitivity, 75% specificity, 74% overall accuracy and 77% area under the curve). We found that apparent diffusion coefficient, fractional anisotropy and radial diffusivity contributed more than other features to the model. Incorporating the patient demographics and clinical features such as age, tumour World Health Organization grade, tumour location, gender and resting motor threshold did not affect the model's performance, revealing that these features were not as effective as microstructural measures. These results shed light on the potential patterns of tumour-related microstructural white matter changes in the prediction of functional deficits. KW - machine learning KW - support vector machine KW - tractography KW - diffusion MRI; KW - corticospinal tract Y1 - 2022 U6 - https://doi.org/10.1093/braincomms/fcac141 SN - 2632-1297 VL - 4 IS - 3 PB - Oxford University Press CY - Oxford ER - TY - BOOK A1 - Kuban, Robert A1 - Rotta, Randolf A1 - Nolte, Jörg A1 - Chromik, Jonas A1 - Beilharz, Jossekin Jakob A1 - Pirl, Lukas A1 - Friedrich, Tobias A1 - Lenzner, Pascal A1 - Weyand, Christopher A1 - Juiz, Carlos A1 - Bermejo, Belen A1 - Sauer, Joao A1 - Coelh, Leandro dos Santos A1 - Najafi, Pejman A1 - Pünter, Wenzel A1 - Cheng, Feng A1 - Meinel, Christoph A1 - Sidorova, Julia A1 - Lundberg, Lars A1 - Vogel, Thomas A1 - Tran, Chinh A1 - Moser, Irene A1 - Grunske, Lars A1 - Elsaid, Mohamed Esameldin Mohamed A1 - Abbas, Hazem M. A1 - Rula, Anisa A1 - Sejdiu, Gezim A1 - Maurino, Andrea A1 - Schmidt, Christopher A1 - Hügle, Johannes A1 - Uflacker, Matthias A1 - Nozza, Debora A1 - Messina, Enza A1 - Hoorn, André van A1 - Frank, Markus A1 - Schulz, Henning A1 - Alhosseini Almodarresi Yasin, Seyed Ali A1 - Nowicki, Marek A1 - Muite, Benson K. A1 - Boysan, Mehmet Can A1 - Bianchi, Federico A1 - Cremaschi, Marco A1 - Moussa, Rim A1 - Abdel-Karim, Benjamin M. A1 - Pfeuffer, Nicolas A1 - Hinz, Oliver A1 - Plauth, Max A1 - Polze, Andreas A1 - Huo, Da A1 - Melo, Gerard de A1 - Mendes Soares, Fábio A1 - Oliveira, Roberto Célio Limão de A1 - Benson, Lawrence A1 - Paul, Fabian A1 - Werling, Christian A1 - Windheuser, Fabian A1 - Stojanovic, Dragan A1 - Djordjevic, Igor A1 - Stojanovic, Natalija A1 - Stojnev Ilic, Aleksandra A1 - Weidmann, Vera A1 - Lowitzki, Leon A1 - Wagner, Markus A1 - Ifa, Abdessatar Ben A1 - Arlos, Patrik A1 - Megia, Ana A1 - Vendrell, Joan A1 - Pfitzner, Bjarne A1 - Redondo, Alberto A1 - Ríos Insua, David A1 - Albert, Justin Amadeus A1 - Zhou, Lin A1 - Arnrich, Bert A1 - Szabó, Ildikó A1 - Fodor, Szabina A1 - Ternai, Katalin A1 - Bhowmik, Rajarshi A1 - Campero Durand, Gabriel A1 - Shevchenko, Pavlo A1 - Malysheva, Milena A1 - Prymak, Ivan A1 - Saake, Gunter ED - Meinel, Christoph ED - Polze, Andreas ED - Beins, Karsten ED - Strotmann, Rolf ED - Seibold, Ulrich ED - Rödszus, Kurt ED - Müller, Jürgen T1 - HPI Future SOC Lab – Proceedings 2019 N2 - The “HPI Future SOC Lab” is a cooperation of the Hasso Plattner Institute (HPI) and industry partners. Its mission is to enable and promote exchange and interaction between the research community and the industry partners. The HPI Future SOC Lab provides researchers with free of charge access to a complete infrastructure of state of the art hard and software. This infrastructure includes components, which might be too expensive for an ordinary research environment, such as servers with up to 64 cores and 2 TB main memory. The offerings address researchers particularly from but not limited to the areas of computer science and business information systems. Main areas of research include cloud computing, parallelization, and In-Memory technologies. This technical report presents results of research projects executed in 2019. Selected projects have presented their results on April 9th and November 12th 2019 at the Future SOC Lab Day events. N2 - Das Future SOC Lab am HPI ist eine Kooperation des Hasso-Plattner-Instituts mit verschiedenen Industriepartnern. Seine Aufgabe ist die Ermöglichung und Förderung des Austausches zwischen Forschungsgemeinschaft und Industrie. Am Lab wird interessierten Wissenschaftlern eine Infrastruktur von neuester Hard- und Software kostenfrei für Forschungszwecke zur Verfügung gestellt. Dazu zählen teilweise noch nicht am Markt verfügbare Technologien, die im normalen Hochschulbereich in der Regel nicht zu finanzieren wären, bspw. Server mit bis zu 64 Cores und 2 TB Hauptspeicher. Diese Angebote richten sich insbesondere an Wissenschaftler in den Gebieten Informatik und Wirtschaftsinformatik. Einige der Schwerpunkte sind Cloud Computing, Parallelisierung und In-Memory Technologien. In diesem Technischen Bericht werden die Ergebnisse der Forschungsprojekte des Jahres 2019 vorgestellt. Ausgewählte Projekte stellten ihre Ergebnisse am 09. April und 12. November 2019 im Rahmen des Future SOC Lab Tags vor. T3 - Technische Berichte des Hasso-Plattner-Instituts für Digital Engineering an der Universität Potsdam - 158 KW - Future SOC Lab KW - research projects KW - multicore architectures KW - in-memory technology KW - cloud computing KW - machine learning KW - artifical intelligence KW - Future SOC Lab KW - Forschungsprojekte KW - Multicore Architekturen KW - In-Memory Technologie KW - Cloud Computing KW - maschinelles Lernen KW - künstliche Intelligenz Y1 - 2023 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:517-opus4-597915 SN - 978-3-86956-564-4 SN - 1613-5652 SN - 2191-1665 IS - 158 PB - Universitätsverlag Potsdam CY - Potsdam ER - TY - JOUR A1 - Kappattanavar, Arpita Mallikarjuna A1 - Hecker, Pascal A1 - Moontaha, Sidratul A1 - Steckhan, Nico A1 - Arnrich, Bert T1 - Food choices after cognitive load BT - an affective computing approach JF - Sensors N2 - Psychology and nutritional science research has highlighted the impact of negative emotions and cognitive load on calorie consumption behaviour using subjective questionnaires. Isolated studies in other domains objectively assess cognitive load without considering its effects on eating behaviour. This study aims to explore the potential for developing an integrated eating behaviour assistant system that incorporates cognitive load factors. Two experimental sessions were conducted using custom-developed experimentation software to induce different stimuli. During these sessions, we collected 30 h of physiological, food consumption, and affective states questionnaires data to automatically detect cognitive load and analyse its effect on food choice. Utilising grid search optimisation and leave-one-subject-out cross-validation, a support vector machine model achieved a mean classification accuracy of 85.12% for the two cognitive load tasks using eight relevant features. Statistical analysis was performed on calorie consumption and questionnaire data. Furthermore, 75% of the subjects with higher negative affect significantly increased consumption of specific foods after high-cognitive-load tasks. These findings offer insights into the intricate relationship between cognitive load, affective states, and food choice, paving the way for an eating behaviour assistant system to manage food choices during cognitive load. Future research should enhance system capabilities and explore real-world applications. KW - cognitive load KW - eating behaviour KW - machine learning KW - physiological signals KW - photoplethysmography KW - electrodermal activity KW - sensors Y1 - 2023 U6 - https://doi.org/10.3390/s23146597 SN - 1424-8220 VL - 23 IS - 14 PB - MDPI CY - Basel ER - TY - JOUR A1 - Hollenstein, Nora A1 - Trondle, Marius A1 - Plomecka, Martyna A1 - Kiegeland, Samuel A1 - Ozyurt, Yilmazcan A1 - Jäger, Lena Ann A1 - Langer, Nicolas T1 - The ZuCo benchmark on cross-subject reading task classification with EEG and eye-tracking data JF - Frontiers in psychology N2 - We present a new machine learning benchmark for reading task classification with the goal of advancing EEG and eye-tracking research at the intersection between computational language processing and cognitive neuroscience. The benchmark task consists of a cross-subject classification to distinguish between two reading paradigms: normal reading and task-specific reading. The data for the benchmark is based on the Zurich Cognitive Language Processing Corpus (ZuCo 2.0), which provides simultaneous eye-tracking and EEG signals from natural reading of English sentences. The training dataset is publicly available, and we present a newly recorded hidden testset. We provide multiple solid baseline methods for this task and discuss future improvements. We release our code and provide an easy-to-use interface to evaluate new approaches with an accompanying public leaderboard: . KW - reading task classification KW - eye-tracking KW - EEG KW - machine learning KW - reading research KW - cross-subject evaluation Y1 - 2023 U6 - https://doi.org/10.3389/fpsyg.2022.1028824 SN - 1664-1078 VL - 13 PB - Frontiers Media CY - Lausanne ER - TY - JOUR A1 - Rosso, Pablo A1 - Nendel, Claas A1 - Gilardi, Nicolas A1 - Udroiu, Cosmin A1 - Chlebowski, Florent T1 - Processing of remote sensing information to retrieve leaf area index in barley BT - a comparison of methods JF - Precision agriculture N2 - Leaf area index (LAI) is a key variable in understanding and modeling crop-environment interactions. With the advent of increasingly higher spatial resolution satellites and sensors mounted on remotely piloted aircrafts (RPAs), the use of remote sensing in precision agriculture is becoming more common. Since also the availability of methods to retrieve LAI from image data have also drastically expanded, it is necessary to test simultaneously as many methods as possible to understand the advantages and disadvantages of each approach. Ground-based LAI data from three years of barley experiments were related to remote sensing information using vegetation indices (VI), machine learning (ML) and radiative transfer models (RTM), to assess the relative accuracy and efficacy of these methods. The optimized soil adjusted vegetation index and a modified version of the Weighted Difference Vegetation Index performed slightly better than any other retrieval method. However, all methods yielded coefficients of determination of around 0.7 to 0.9. The best performing machine learning algorithms achieved higher accuracies when four Sentinel-2 bands instead of 12 were used. Also, the good performance of VIs and the satisfactory performance of the 4-band RTM, strongly support the synergistic use of satellites and RPAs in precision agriculture. One of the methods used, Sen2-Agri, an open source ML-RTM-based operational system, was also able to accurately retrieve LAI, although it is restricted to Sentinel-2 and Landsat data. This study shows the benefits of testing simultaneously a broad range of retrieval methods to monitor crops for precision agriculture. KW - leaf area index KW - vegetation indices KW - machine learning KW - radiative transfer models Y1 - 2022 U6 - https://doi.org/10.1007/s11119-022-09893-4 SN - 1385-2256 SN - 1573-1618 VL - 23 IS - 4 SP - 1449 EP - 1472 PB - Springer CY - Dordrecht ER - TY - JOUR A1 - Baumgart, Lene A1 - Boos, Pauline A1 - Eckstein, Bernd T1 - Datafication and algorithmic contingency BT - how agile organisations deal with technical systems JF - Work organisation, labour & globalisation N2 - In the context of persistent images of self-perpetuated technologies, we discuss the interplay of digital technologies and organisational dynamics against the backdrop of systems theory. Building on the case of an international corporation that, during an agile reorganisation, introduced an AI-based personnel management platform, we show how technical systems produce a form of algorithmic contingency that subsequently leads to the emergence of formal and informal interaction systems. Using the concept of datafication, we explain how these interactions are barriers to the self-perpetuation of data-based decision-making, making it possible to take into consideration further decision factors and complementing the output of the platform. The research was carried out within the scope of the research project ‘Organisational Implications of Digitalisation: The Development of (Post-)Bureaucratic Organisational Structures in the Context of Digital Transformation’ funded by the German Research Foundation (DFG). KW - digitalisation KW - datafication KW - organisation KW - agile KW - technical system KW - systems theory KW - interaction KW - algorithmic contingency KW - machine learning KW - platform Y1 - 2023 U6 - https://doi.org/10.13169/workorgalaboglob.17.1.0061 SN - 1745-641X SN - 1745-6428 VL - 17 IS - 1 SP - 61 EP - 73 PB - Pluto Journals CY - London ER - TY - RPRT A1 - Andres, Maximilian A1 - Bruttel, Lisa T1 - Communicating Cartel Intentions T2 - CEPA Discussion Papers N2 - While the economic harm of cartels is caused by their price-increasing effect, sanctioning by courts rather targets at the underlying process of firms reaching a price-fixing agreement. This paper provides experimental evidence on the question whether such sanctioning meets the economic target, i.e., whether evidence of a collusive meeting of the firms and of the content of their communication reliably predicts subsequent prices. We find that already the mere mutual agreement to meet predicts a strong increase in prices. Conversely, express distancing from communication completely nullifies its otherwise price-increasing effect. Using machine learning, we show that communication only increases prices if it is very explicit about how the cartel plans to behave. T3 - CEPA Discussion Papers - 77 KW - cartel KW - collusion KW - communication KW - machine learning KW - experiment Y1 - 2024 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:517-opus4-638469 SN - 2628-653X IS - 77 ER - TY - GEN A1 - Panzer, Marcel A1 - Bender, Benedict A1 - Gronau, Norbert T1 - Deep reinforcement learning in production planning and control BT - A systematic literature review T2 - Zweitveröffentlichungen der Universität Potsdam : Wirtschafts- und Sozialwissenschaftliche Reihe N2 - Increasingly fast development cycles and individualized products pose major challenges for today's smart production systems in times of industry 4.0. The systems must be flexible and continuously adapt to changing conditions while still guaranteeing high throughputs and robustness against external disruptions. Deep reinforcement learning (RL) algorithms, which already reached impressive success with Google DeepMind's AlphaGo, are increasingly transferred to production systems to meet related requirements. Unlike supervised and unsupervised machine learning techniques, deep RL algorithms learn based on recently collected sensorand process-data in direct interaction with the environment and are able to perform decisions in real-time. As such, deep RL algorithms seem promising given their potential to provide decision support in complex environments, as production systems, and simultaneously adapt to changing circumstances. While different use-cases for deep RL emerged, a structured overview and integration of findings on their application are missing. To address this gap, this contribution provides a systematic literature review of existing deep RL applications in the field of production planning and control as well as production logistics. From a performance perspective, it became evident that deep RL can beat heuristics significantly in their overall performance and provides superior solutions to various industrial use-cases. Nevertheless, safety and reliability concerns must be overcome before the widespread use of deep RL is possible which presumes more intensive testing of deep RL in real world applications besides the already ongoing intensive simulations. T3 - Zweitveröffentlichungen der Universität Potsdam : Wirtschafts- und Sozialwissenschaftliche Reihe - 198 KW - deep reinforcement learning KW - machine learning KW - production planning KW - production control KW - systematic literature review Y1 - 2021 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:517-opus4-605722 SN - 2701-6277 SN - 1867-5808 ER - TY - GEN A1 - Panzer, Marcel A1 - Bender, Benedict A1 - Gronau, Norbert T1 - Neural agent-based production planning and control BT - an architectural review T2 - Zweitveröffentlichungen der Universität Potsdam : Wirtschafts- und Sozialwissenschaftliche Reihe N2 - Nowadays, production planning and control must cope with mass customization, increased fluctuations in demand, and high competition pressures. Despite prevailing market risks, planning accuracy and increased adaptability in the event of disruptions or failures must be ensured, while simultaneously optimizing key process indicators. To manage that complex task, neural networks that can process large quantities of high-dimensional data in real time have been widely adopted in recent years. Although these are already extensively deployed in production systems, a systematic review of applications and implemented agent embeddings and architectures has not yet been conducted. The main contribution of this paper is to provide researchers and practitioners with an overview of applications and applied embeddings and to motivate further research in neural agent-based production. Findings indicate that neural agents are not only deployed in diverse applications, but are also increasingly implemented in multi-agent environments or in combination with conventional methods — leveraging performances compared to benchmarks and reducing dependence on human experience. This not only implies a more sophisticated focus on distributed production resources, but also broadening the perspective from a local to a global scale. Nevertheless, future research must further increase scalability and reproducibility to guarantee a simplified transfer of results to reality. T3 - Zweitveröffentlichungen der Universität Potsdam : Wirtschafts- und Sozialwissenschaftliche Reihe - 172 KW - production planning and control KW - machine learning KW - neural networks KW - systematic literature review KW - taxonomy Y1 - 2022 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:517-opus4-604777 SN - 1867-5808 ER - TY - JOUR A1 - Panzer, Marcel A1 - Bender, Benedict A1 - Gronau, Norbert T1 - Neural agent-based production planning and control BT - an architectural review JF - Journal of Manufacturing Systems N2 - Nowadays, production planning and control must cope with mass customization, increased fluctuations in demand, and high competition pressures. Despite prevailing market risks, planning accuracy and increased adaptability in the event of disruptions or failures must be ensured, while simultaneously optimizing key process indicators. To manage that complex task, neural networks that can process large quantities of high-dimensional data in real time have been widely adopted in recent years. Although these are already extensively deployed in production systems, a systematic review of applications and implemented agent embeddings and architectures has not yet been conducted. The main contribution of this paper is to provide researchers and practitioners with an overview of applications and applied embeddings and to motivate further research in neural agent-based production. Findings indicate that neural agents are not only deployed in diverse applications, but are also increasingly implemented in multi-agent environments or in combination with conventional methods — leveraging performances compared to benchmarks and reducing dependence on human experience. This not only implies a more sophisticated focus on distributed production resources, but also broadening the perspective from a local to a global scale. Nevertheless, future research must further increase scalability and reproducibility to guarantee a simplified transfer of results to reality. KW - production planning and control KW - machine learning KW - neural networks KW - systematic literature review KW - taxonomy Y1 - 2022 U6 - https://doi.org/10.1016/j.jmsy.2022.10.019 SN - 0278-6125 SN - 1878-6642 VL - 65 SP - 743 EP - 766 PB - Elsevier CY - Amsterdam ER - TY - THES A1 - Lilienkamp, Henning T1 - Enhanced computational approaches for data-driven characterization of earthquake ground motion and rapid earthquake impact assessment T1 - Fortgeschrittene Berechnungsansätze für die datengestützte Charakterisierung von Erdbeben-Bodenbewegungen und die schnelle Einschätzung von Erdbebenauswirkungen N2 - Rapidly growing seismic and macroseismic databases and simplified access to advanced machine learning methods have in recent years opened up vast opportunities to address challenges in engineering and strong motion seismology from novel, datacentric perspectives. In this thesis, I explore the opportunities of such perspectives for the tasks of ground motion modeling and rapid earthquake impact assessment, tasks with major implications for long-term earthquake disaster mitigation. In my first study, I utilize the rich strong motion database from the Kanto basin, Japan, and apply the U-Net artificial neural network architecture to develop a deep learning based ground motion model. The operational prototype provides statistical estimates of expected ground shaking, given descriptions of a specific earthquake source, wave propagation paths, and geophysical site conditions. The U-Net interprets ground motion data in its spatial context, potentially taking into account, for example, the geological properties in the vicinity of observation sites. Predictions of ground motion intensity are thereby calibrated to individual observation sites and earthquake locations. The second study addresses the explicit incorporation of rupture forward directivity into ground motion modeling. Incorporation of this phenomenon, causing strong, pulse like ground shaking in the vicinity of earthquake sources, is usually associated with an intolerable increase in computational demand during probabilistic seismic hazard analysis (PSHA) calculations. I suggest an approach in which I utilize an artificial neural network to efficiently approximate the average, directivity-related adjustment to ground motion predictions for earthquake ruptures from the 2022 New Zealand National Seismic Hazard Model. The practical implementation in an actual PSHA calculation demonstrates the efficiency and operational readiness of my model. In a follow-up study, I present a proof of concept for an alternative strategy in which I target the generalizing applicability to ruptures other than those from the New Zealand National Seismic Hazard Model. In the third study, I address the usability of pseudo-intensity reports obtained from macroseismic observations by non-expert citizens for rapid impact assessment. I demonstrate that the statistical properties of pseudo-intensity collections describing the intensity of shaking are correlated with the societal impact of earthquakes. In a second step, I develop a probabilistic model that, within minutes of an event, quantifies the probability of an earthquake to cause considerable societal impact. Under certain conditions, such a quick and preliminary method might be useful to support decision makers in their efforts to organize auxiliary measures for earthquake disaster response while results from more elaborate impact assessment frameworks are not yet available. The application of machine learning methods to datasets that only partially reveal characteristics of Big Data, qualify the majority of results obtained in this thesis as explorative insights rather than ready-to-use solutions to real world problems. The practical usefulness of this work will be better assessed in the future by applying the approaches developed to growing and increasingly complex data sets. N2 - Das rapide Wachstum seismischer und makroseismischer Datenbanken und der vereinfachte Zugang zu fortschrittlichen Methoden aus dem Bereich des maschinellen Lernens haben in den letzen Jahren die datenfokussierte Betrachtung von Fragestellungen in der Seismologie ermöglicht. In dieser Arbeit erforsche ich das Potenzial solcher Betrachtungsweisen im Hinblick auf die Modellierung erdbebenbedingter Bodenerschütterungen und der raschen Einschätzung von gesellschaftlichen Erdbebenauswirkungen, Disziplinen von erheblicher Bedeutung für den langfristigen Erdbebenkatastrophenschutz in seismisch aktiven Regionen. In meiner ersten Studie nutze ich die Vielzahl an Bodenbewegungsdaten aus der Kanto Region in Japan, sowie eine spezielle neuronale Netzwerkarchitektur (U-Net) um ein Bodenbewegungsmodell zu entwickeln. Der einsatzbereite Prototyp liefert auf Basis der Charakterisierung von Erdbebenherden, Wellenausbreitungspfaden und Bodenbeschaffenheiten statistische Schätzungen der zu erwartenden Bodenerschütterungen. Das U-Net interpretiert Bodenbewegungsdaten im räumlichen Kontext, sodass etwa die geologischen Beschaffenheiten in der Umgebung von Messstationen mit einbezogen werden können. Auch die absoluten Koordinaten von Erdbebenherden und Messstationen werden berücksichtigt. Die zweite Studie behandelt die explizite Berücksichtigung richtungsabhängiger Verstärkungseffekte in der Bodenbewegungsmodellierung. Obwohl solche Effekte starke, impulsartige Erschütterungen in der Nähe von Erdbebenherden erzeugen, die eine erhebliche seismische Beanspruchung von Gebäuden darstellen, wird deren explizite Modellierung in der seismischen Gefährdungsabschätzung aufgrund des nicht vertretbaren Rechenaufwandes ausgelassen. Mit meinem, auf einem neuronalen Netzwerk basierenden, Ansatz schlage ich eine Methode vor, umdieses Vorhaben effizient für Erdbebenszenarien aus dem neuseeländischen seismischen Gefährdungsmodell für 2022 (NSHM) umzusetzen. Die Implementierung in einer seismischen Gefährdungsrechnung unterstreicht die Praktikabilität meines Modells. In einer anschließenden Machbarkeitsstudie untersuche ich einen alternativen Ansatz der auf die Anwendbarkeit auf beliebige Erdbebeszenarien abzielt. Die abschließende dritte Studie befasst sich mit dem potenziellen Nutzen der von makroseismischen Beobachtungen abgeleiteten pseudo-Erschütterungsintensitäten für die rasche Abschätzung von gesellschaftlichen Erdbebenauswirkungen. Ich zeige, dass sich aus den Merkmalen solcher Daten Schlussfolgerungen über die gesellschaftlichen Folgen eines Erdbebens ableiten lassen. Basierend darauf formuliere ich ein statistisches Modell, welches innerhalb weniger Minuten nach einem Erdbeben die Wahrscheinlichkeit für das Auftreten beachtlicher gesellschaftlicher Auswirkungen liefert. Ich komme zu dem Schluss, dass ein solches Modell, unter bestimmten Bedingungen, hilfreich sein könnte, um EntscheidungsträgerInnen in ihren Bestrebungen Hilfsmaßnahmen zu organisieren zu unterstützen. Die Anwendung von Methoden des maschinellen Lernens auf Datensätze die sich nur begrenzt als Big Data charakterisieren lassen, qualifizieren die Mehrheit der Ergebnisse dieser Arbeit als explorative Einblicke und weniger als einsatzbereite Lösungen für praktische Fragestellungen. Der praktische Nutzen dieser Arbeit wird sich in erst in Zukunft an der Anwendung der erarbeiteten Ansätze auf wachsende und zunehmend komplexe Datensätze final abschätzen lassen. KW - seismology KW - machine learning KW - deep learning KW - ground motion modeling KW - seismic hazard KW - rapid earthquake impact assessment KW - geophysics KW - Deep Learning KW - Geophysik KW - Bodenbewegungsmodellierung KW - maschinelles Lernen KW - schnelle Einschätzung von Erdbebenauswirkungen KW - seismische Gefährdung KW - Seismologie Y1 - 2024 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:517-opus4-631954 ER - TY - JOUR A1 - Hampf, Anna A1 - Nendel, Claas A1 - Strey, Simone A1 - Strey, Robert T1 - Biotic yield losses in the Southern Amazon, Brazil BT - making use of smartphone-assisted plant disease diagnosis data JF - Frontiers in plant science : FPLS N2 - Pathogens and animal pests (P&A) are a major threat to global food security as they directly affect the quantity and quality of food. The Southern Amazon, Brazil's largest domestic region for soybean, maize and cotton production, is particularly vulnerable to the outbreak of P&A due to its (sub)tropical climate and intensive farming systems. However, little is known about the spatial distribution of P&A and the related yield losses. Machine learning approaches for the automated recognition of plant diseases can help to overcome this research gap. The main objectives of this study are to (1) evaluate the performance of Convolutional Neural Networks (ConvNets) in classifying P&A, (2) map the spatial distribution of P&A in the Southern Amazon, and (3) quantify perceived yield and economic losses for the main soybean and maize P&A. The objectives were addressed by making use of data collected with the smartphone application Plantix. The core of the app's functioning is the automated recognition of plant diseases via ConvNets. Data on expected yield losses were gathered through a short survey included in an "expert" version of the application, which was distributed among agronomists. Between 2016 and 2020, Plantix users collected approximately 78,000 georeferenced P&A images in the Southern Amazon. The study results indicate a high performance of the trained ConvNets in classifying 420 different crop-disease combinations. Spatial distribution maps and expert-based yield loss estimates indicate that maize rust, bacterial stalk rot and the fall armyworm are among the most severe maize P&A, whereas soybean is mainly affected by P&A like anthracnose, downy mildew, frogeye leaf spot, stink bugs and brown spot. Perceived soybean and maize yield losses amount to 12 and 16%, respectively, resulting in annual yield losses of approximately 3.75 million tonnes for each crop and economic losses of US$2 billion for both crops together. The high level of accuracy of the trained ConvNets, when paired with widespread use from following a citizen-science approach, results in a data source that will shed new light on yield loss estimates, e.g., for the analysis of yield gaps and the development of measures to minimise them. KW - plant pathology KW - animal pests KW - pathogens KW - machine learning KW - digital KW - image processing KW - disease diagnosis KW - crowdsourcing KW - crop losses Y1 - 2021 U6 - https://doi.org/10.3389/fpls.2021.621168 SN - 1664-462X VL - 12 PB - Frontiers Media CY - Lausanne ER - TY - JOUR A1 - Andres, Maximilian A1 - Bruttel, Lisa A1 - Friedrichsen, Jana T1 - How communication makes the difference between a cartel and tacit collusion BT - a machine learning approach JF - European economic review N2 - This paper sheds new light on the role of communication for cartel formation. Using machine learning to evaluate free-form chat communication among firms in a laboratory experiment, we identify typical communication patterns for both explicit cartel formation and indirect attempts to collude tacitly. We document that firms are less likely to communicate explicitly about price fixing and more likely to use indirect messages when sanctioning institutions are present. This effect of sanctions on communication reinforces the direct cartel-deterring effect of sanctions as collusion is more difficult to reach and sustain without an explicit agreement. Indirect messages have no, or even a negative, effect on prices. KW - cartel KW - collusion KW - communication KW - machine learning KW - experiment Y1 - 2023 U6 - https://doi.org/10.1016/j.euroecorev.2022.104331 SN - 0014-2921 SN - 1873-572X VL - 152 SP - 1 EP - 18 PB - Elsevier CY - Amsterdam ER - TY - JOUR A1 - Chen, Junchao A1 - Lange, Thomas A1 - Andjelkovic, Marko A1 - Simevski, Aleksandar A1 - Lu, Li A1 - Krstić, Miloš T1 - Solar particle event and single event upset prediction from SRAM-based monitor and supervised machine learning JF - IEEE transactions on emerging topics in computing / IEEE Computer Society, Institute of Electrical and Electronics Engineers N2 - The intensity of cosmic radiation may differ over five orders of magnitude within a few hours or days during the Solar Particle Events (SPEs), thus increasing for several orders of magnitude the probability of Single Event Upsets (SEUs) in space-borne electronic systems. Therefore, it is vital to enable the early detection of the SEU rate changes in order to ensure timely activation of dynamic radiation hardening measures. In this paper, an embedded approach for the prediction of SPEs and SRAM SEU rate is presented. The proposed solution combines the real-time SRAM-based SEU monitor, the offline-trained machine learning model and online learning algorithm for the prediction. With respect to the state-of-the-art, our solution brings the following benefits: (1) Use of existing on-chip data storage SRAM as a particle detector, thus minimizing the hardware and power overhead, (2) Prediction of SRAM SEU rate one hour in advance, with the fine-grained hourly tracking of SEU variations during SPEs as well as under normal conditions, (3) Online optimization of the prediction model for enhancing the prediction accuracy during run-time, (4) Negligible cost of hardware accelerator design for the implementation of selected machine learning model and online learning algorithm. The proposed design is intended for a highly dependable and self-adaptive multiprocessing system employed in space applications, allowing to trigger the radiation mitigation mechanisms before the onset of high radiation levels. KW - Machine learning KW - Single event upsets KW - Random access memory KW - monitoring KW - machine learning algorithms KW - predictive models KW - space missions KW - solar particle event KW - single event upset KW - machine learning KW - online learning KW - hardware accelerator KW - reliability KW - self-adaptive multiprocessing system Y1 - 2022 U6 - https://doi.org/10.1109/TETC.2022.3147376 SN - 2168-6750 VL - 10 IS - 2 SP - 564 EP - 580 PB - Institute of Electrical and Electronics Engineers CY - [New York, NY] ER - TY - JOUR A1 - Ghafarian, Fatemeh A1 - Wieland, Ralf A1 - Lüttschwager, Dietmar A1 - Nendel, Claas T1 - Application of extreme gradient boosting and Shapley Additive explanations to predict temperature regimes inside forests from standard open-field meteorological data JF - Environmental modelling & software with environment data news N2 - Forest microclimate can buffer biotic responses to summer heat waves, which are expected to become more extreme under climate warming. Prediction of forest microclimate is limited because meteorological observation standards seldom include situations inside forests. We use eXtreme Gradient Boosting - a Machine Learning technique - to predict the microclimate of forest sites in Brandenburg, Germany, using seasonal data comprising weather features. The analysis was amended by applying a SHapley Additive explanation to show the interaction effect of variables and individualised feature attributions. We evaluate model performance in comparison to artificial neural networks, random forest, support vector machine, and multi-linear regression. After implementing a feature selection, an ensemble approach was applied to combine individual models for each forest and improve robustness over a given single prediction model. The resulting model can be applied to translate climate change scenarios into temperatures inside forests to assess temperature-related ecosystem services provided by forests. KW - cooling effect KW - machine learning KW - ensemble method KW - ecosystem services Y1 - 2022 U6 - https://doi.org/10.1016/j.envsoft.2022.105466 SN - 1364-8152 SN - 1873-6726 VL - 156 PB - Elsevier CY - Oxford ER - TY - RPRT A1 - Andres, Maximilian T1 - Equilibrium selection in infinitely repeated games with communication T2 - CEPA Discussion Papers N2 - The present paper proposes a novel approach for equilibrium selection in the infinitely repeated prisoner’s dilemma where players can communicate before choosing their strategies. This approach yields a critical discount factor that makes different predictions for cooperation than the usually considered sub-game perfect or risk dominance critical discount factors. In laboratory experiments, we find that our factor is useful for predicting cooperation. For payoff changes where the usually considered factors and our factor make different predictions, the observed cooperation is consistent with the predictions based on our factor. T3 - CEPA Discussion Papers - 75 KW - cooperation KW - communication KW - infinitely repeated game KW - machine learning Y1 - 2024 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:517-opus4-631800 SN - 2628-653X IS - 75 ER - TY - JOUR A1 - Kühn, Daniela A1 - Hainzl, Sebastian A1 - Dahm, Torsten A1 - Richter, Gudrun A1 - Vera Rodriguez, Ismael T1 - A review of source models to further the understanding of the seismicity of the Groningen field JF - Netherlands journal of geosciences : NJG N2 - The occurrence of felt earthquakes due to gas production in Groningen has initiated numerous studies and model attempts to understand and quantify induced seismicity in this region. The whole bandwidth of available models spans the range from fully deterministic models to purely empirical and stochastic models. In this article, we summarise the most important model approaches, describing their main achievements and limitations. In addition, we discuss remaining open questions and potential future directions of development. KW - deterministic KW - empirical KW - hybrid KW - machine learning KW - seismicity model Y1 - 2022 U6 - https://doi.org/10.1017/njg.2022.7 SN - 0016-7746 SN - 1573-9708 VL - 101 PB - Cambridge Univ. Press CY - Cambridge ER - TY - JOUR A1 - Wilksch, Moritz A1 - Abramova, Olga T1 - PyFin-sentiment BT - towards a machine-learning-based model for deriving sentiment from financial tweets JF - International journal of information management data insights N2 - Responding to the poor performance of generic automated sentiment analysis solutions on domain-specific texts, we collect a dataset of 10,000 tweets discussing the topics of finance and investing. We manually assign each tweet its market sentiment, i.e., the investor’s anticipation of a stock’s future return. Using this data, we show that all existing sentiment models trained on adjacent domains struggle with accurate market sentiment analysis due to the task’s specialized vocabulary. Consequently, we design, train, and deploy our own sentiment model. It outperforms all previous models (VADER, NTUSD-Fin, FinBERT, TwitterRoBERTa) when evaluated on Twitter posts. On posts from a different platform, our model performs on par with BERT-based large language models. We achieve this result at a fraction of the training and inference costs due to the model’s simple design. We publish the artifact as a python library to facilitate its use by future researchers and practitioners. KW - sentiment analysis KW - financial market sentiment KW - opinion mining KW - machine learning KW - deep learning Y1 - 2023 U6 - https://doi.org/10.1016/j.jjimei.2023.100171 SN - 2667-0968 VL - 3 IS - 1 PB - Elsevier CY - Amsterdam ER - TY - JOUR A1 - Brandes, Stefanie A1 - Sicks, Florian A1 - Berger, Anne T1 - Behaviour classification on giraffes (Giraffa camelopardalis) using machine learning algorithms on triaxial acceleration data of two commonly used GPS devices and its possible application for their management and conservation JF - Sensors N2 - Averting today's loss of biodiversity and ecosystem services can be achieved through conservation efforts, especially of keystone species. Giraffes (Giraffa camelopardalis) play an important role in sustaining Africa's ecosystems, but are 'vulnerable' according to the IUCN Red List since 2016. Monitoring an animal's behavior in the wild helps to develop and assess their conservation management. One mechanism for remote tracking of wildlife behavior is to attach accelerometers to animals to record their body movement. We tested two different commercially available high-resolution accelerometers, e-obs and Africa Wildlife Tracking (AWT), attached to the top of the heads of three captive giraffes and analyzed the accuracy of automatic behavior classifications, focused on the Random Forests algorithm. For both accelerometers, behaviors of lower variety in head and neck movements could be better predicted (i.e., feeding above eye level, mean prediction accuracy e-obs/AWT: 97.6%/99.7%; drinking: 96.7%/97.0%) than those with a higher variety of body postures (such as standing: 90.7-91.0%/75.2-76.7%; rumination: 89.6-91.6%/53.5-86.5%). Nonetheless both devices come with limitations and especially the AWT needs technological adaptations before applying it on animals in the wild. Nevertheless, looking at the prediction results, both are promising accelerometers for behavioral classification of giraffes. Therefore, these devices when applied to free-ranging animals, in combination with GPS tracking, can contribute greatly to the conservation of giraffes. KW - giraffe KW - triaxial acceleration KW - machine learning KW - random forests KW - behavior classification KW - giraffe conservation Y1 - 2021 U6 - https://doi.org/10.3390/s21062229 SN - 1424-8220 VL - 21 IS - 6 PB - MDPI CY - Basel ER - TY - JOUR A1 - Adnan, Hassan Sami A1 - Srsic, Amanda A1 - Venticich, Pete Milos A1 - Townend, David M.R. T1 - Using AI for mental health analysis and prediction in school surveys JF - European journal of public health N2 - Background: Childhood and adolescence are critical stages of life for mental health and well-being. Schools are a key setting for mental health promotion and illness prevention. One in five children and adolescents have a mental disorder, about half of mental disorders beginning before the age of 14. Beneficial and explainable artificial intelligence can replace current paper- based and online approaches to school mental health surveys. This can enhance data acquisition, interoperability, data driven analysis, trust and compliance. This paper presents a model for using chatbots for non-obtrusive data collection and supervised machine learning models for data analysis; and discusses ethical considerations pertaining to the use of these models. Methods: For data acquisition, the proposed model uses chatbots which interact with students. The conversation log acts as the source of raw data for the machine learning. Pre-processing of the data is automated by filtering for keywords and phrases. Existing survey results, obtained through current paper-based data collection methods, are evaluated by domain experts (health professionals). These can be used to create a test dataset to validate the machine learning models. Supervised learning can then be deployed to classify specific behaviour and mental health patterns. Results: We present a model that can be used to improve upon current paper-based data collection and manual data analysis methods. An open-source GitHub repository contains necessary tools and components of this model. Privacy is respected through rigorous observance of confidentiality and data protection requirements. Critical reflection on these ethics and law aspects is included in the project. Conclusions: This model strengthens mental health surveillance in schools. The same tools and components could be applied to other public health data. Future extensions of this model could also incorporate unsupervised learning to find clusters and patterns of unknown effects. KW - ethics KW - artificial intelligence KW - adolescent KW - child KW - confidentiality KW - health personnel KW - mental disorders KW - mental health KW - personal satisfaction KW - privacy KW - school (environment) KW - statutes and laws KW - public health medicine KW - surveillance KW - medical KW - prevention KW - datasets KW - machine learning KW - supervised machine learning KW - data analysis Y1 - 2020 U6 - https://doi.org/10.1093/eurpub/ckaa165.336 SN - 1101-1262 SN - 1464-360X VL - 30 SP - V125 EP - V125 PB - Oxford Univ. Press CY - Oxford [u.a.] ER - TY - JOUR A1 - Wulff, Peter A1 - Mientus, Lukas A1 - Nowak, Anna A1 - Borowski, Andreas T1 - KI-basierte Auswertung von schriftlichen Unterrichtsreflexionen im Fach Physik und automatisierte Rückmeldung JF - PSI-Potsdam: Ergebnisbericht zu den Aktivitäten im Rahmen der Qualitätsoffensive Lehrerbildung (2019-2023) (Potsdamer Beiträge zur Lehrerbildung und Bildungsforschung ; 3) N2 - Für die Entwicklung professioneller Handlungskompetenzen angehender Lehrkräfte stellt die Unterrichtsreflexion ein wichtiges Instrument dar, um Theoriewissen und Praxiserfahrungen in Beziehung zu setzen. Die Auswertung von Unterrichtsreflexionen und eine entsprechende Rückmeldung stellt Forschende und Dozierende allerdings vor praktische wie theoretische Herausforderungen. Im Kontext der Forschung zu Künstlicher Intelligenz (KI) entwickelte Methoden bieten hier neue Potenziale. Der Beitrag stellt überblicksartig zwei Teilstudien vor, die mit Hilfe von KI-Methoden wie dem maschinellen Lernen untersuchen, inwieweit eine Auswertung von Unterrichtsreflexionen angehender Physiklehrkräfte auf Basis eines theoretisch abgeleiteten Reflexionsmodells und die automatisierte Rückmeldung hierzu möglich sind. Dabei wurden unterschiedliche Ansätze des maschinellen Lernens verwendet, um modellbasierte Klassifikation und Exploration von Themen in Unterrichtsreflexionen umzusetzen. Die Genauigkeit der Ergebnisse wurde vor allem durch sog. Große Sprachmodelle gesteigert, die auch den Transfer auf andere Standorte und Fächer ermöglichen. Für die fachdidaktische Forschung bedeuten sie jedoch wiederum neue Herausforderungen, wie etwa systematische Verzerrungen und Intransparenz von Entscheidungen. Dennoch empfehlen wir, die Potenziale der KI-basierten Methoden gründlicher zu erforschen und konsequent in der Praxis (etwa in Form von Webanwendungen) zu implementieren. N2 - For the development of professional competencies in pre-service teachers, reflection on teaching experiences is proposed as an important tool to link theoretical knowledge and practice. However, evaluating reflections and providing appropriate feedback poses challenges of both theoretical and practical nature to researchers and educators. Methods associated with artificial intelligence research offer new potentials to discover patterns in complex datasets like reflections, as well as to evaluate these automatically and create feedback. In this article, we provide an overview of two sub-studies that investigate, using artificial intelligence methods such as machine learning, to what extent an evaluation of reflections of pre-service physics teachers based on a theoretically derived reflection model and automated feedback are possible. Across the sub-studies, different machine learning approaches were used to implement model-based classification and exploration of topics in reflections. Large language models in particular increase the accuracy of the results and allow for transfer to other locations and disciplines. However, entirely new challenges arise for educational research in relation to large language models, such as systematic biases and lack of transparency in decisions. Despite these uncertainties, we recommend further exploring the potentials of artificial intelligence-based methods and implementing them consistently in practice (for example, in the form of web applications). KW - Künstliche Intelligenz KW - Maschinelles Lernen KW - Natural Language Processing KW - Reflexion KW - Professionalisierung KW - artificial intelligence KW - machine learning KW - natural language processing KW - reflexion KW - professionalization Y1 - 2023 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:517-opus4-616363 SN - 978-3-86956-568-2 SN - 2626-3556 SN - 2626-4722 IS - 3 SP - 103 EP - 115 PB - Universitätsverlag Potsdam CY - Potsdam ER - TY - THES A1 - Najafi, Pejman T1 - Leveraging data science & engineering for advanced security operations T1 - Der Einsatz von Data Science & Engineering für fortschrittliche Security Operations N2 - The Security Operations Center (SOC) represents a specialized unit responsible for managing security within enterprises. To aid in its responsibilities, the SOC relies heavily on a Security Information and Event Management (SIEM) system that functions as a centralized repository for all security-related data, providing a comprehensive view of the organization's security posture. Due to the ability to offer such insights, SIEMS are considered indispensable tools facilitating SOC functions, such as monitoring, threat detection, and incident response. Despite advancements in big data architectures and analytics, most SIEMs fall short of keeping pace. Architecturally, they function merely as log search engines, lacking the support for distributed large-scale analytics. Analytically, they rely on rule-based correlation, neglecting the adoption of more advanced data science and machine learning techniques. This thesis first proposes a blueprint for next-generation SIEM systems that emphasize distributed processing and multi-layered storage to enable data mining at a big data scale. Next, with the architectural support, it introduces two data mining approaches for advanced threat detection as part of SOC operations. First, a novel graph mining technique that formulates threat detection within the SIEM system as a large-scale graph mining and inference problem, built on the principles of guilt-by-association and exempt-by-reputation. The approach entails the construction of a Heterogeneous Information Network (HIN) that models shared characteristics and associations among entities extracted from SIEM-related events/logs. Thereon, a novel graph-based inference algorithm is used to infer a node's maliciousness score based on its associations with other entities in the HIN. Second, an innovative outlier detection technique that imitates a SOC analyst's reasoning process to find anomalies/outliers. The approach emphasizes explainability and simplicity, achieved by combining the output of simple context-aware univariate submodels that calculate an outlier score for each entry. Both approaches were tested in academic and real-world settings, demonstrating high performance when compared to other algorithms as well as practicality alongside a large enterprise's SIEM system. This thesis establishes the foundation for next-generation SIEM systems that can enhance today's SOCs and facilitate the transition from human-centric to data-driven security operations. N2 - In einem Security Operations Center (SOC) werden alle sicherheitsrelevanten Prozesse, Daten und Personen einer Organisation zusammengefasst. Das Herzstück des SOCs ist ein Security Information and Event Management (SIEM)-System, welches als zentraler Speicher aller sicherheitsrelevanten Daten fungiert und einen Überblick über die Sicherheitslage einer Organisation geben kann. SIEM-Systeme sind unverzichtbare Werkzeuge für viele SOC-Funktionen wie Monitoring, Threat Detection und Incident Response. Trotz der Fortschritte bei Big-Data-Architekturen und -Analysen können die meisten SIEMs nicht mithalten. Sie fungieren nur als Protokollsuchmaschine und unterstützen keine verteilte Data Mining und Machine Learning. In dieser Arbeit wird zunächst eine Blaupause für die nächste Generation von SIEM-Systemen vorgestellt, welche Daten verteilt, verarbeitet und in mehreren Schichten speichert, damit auch Data Mining im großen Stil zu ermöglichen. Zudem werden zwei Data Mining-Ansätze vorgeschlagen, mit denen auch anspruchsvolle Bedrohungen erkannt werden können. Der erste Ansatz ist eine neue Graph-Mining-Technik, bei der SIEM-Daten als Graph strukturiert werden und Reputationsinferenz mithilfe der Prinzipien guiltby-association (Kontaktschuld) und exempt-by-reputation (Reputationsbefreiung) implementiert wird. Der Ansatz nutzt ein heterogenes Informationsnetzwerk (HIN), welches gemeinsame Eigenschaften und Assoziationen zwischen Entitäten aus Event Logs verknüpft. Des Weiteren ermöglicht ein neuer Inferenzalgorithmus die Bestimmung der Schädlichkeit eines Kontos anhand seiner Verbindungen zu anderen Entitäten im HIN. Der zweite Ansatz ist eine innovative Methode zur Erkennung von Ausreißern, die den Entscheidungsprozess eines SOC-Analysten imitiert. Diese Methode ist besonders einfach und interpretierbar, da sie einzelne univariate Teilmodelle kombiniert, die sich jeweils auf eine kontextualisierte Eigenschaft einer Entität beziehen. Beide Ansätze wurden sowohl akademisch als auch in der Praxis getestet und haben im Vergleich mit anderen Methoden auch in großen Unternehmen eine hohe Qualität bewiesen. Diese Arbeit bildet die Grundlage für die nächste Generation von SIEM-Systemen, welche den Übergang von einer personalzentrischen zu einer datenzentrischen Perspektive auf SOCs ermöglichen. KW - cybersecurity KW - endpoint security KW - threat detection KW - intrusion detection KW - apt KW - advanced threats KW - advanced persistent threat KW - zero-day KW - security analytics KW - data-driven KW - data mining KW - data science KW - anomaly detection KW - outlier detection KW - graph mining KW - graph inference KW - machine learning KW - Advanced Persistent Threats KW - fortschrittliche Angriffe KW - Anomalieerkennung KW - APT KW - Cyber-Sicherheit KW - Data-Mining KW - Data-Science KW - datengetrieben KW - Endpunktsicherheit KW - Graphableitung KW - Graph-Mining KW - Einbruchserkennung KW - Machine-Learning KW - Ausreißererkennung KW - Sicherheitsanalyse KW - Bedrohungserkennung KW - 0-day Y1 - 2023 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:517-opus4-612257 ER - TY - THES A1 - Smirnov, Artem T1 - Understanding the dynamics of the near-earth space environment utilizing long-term satellite observations T1 - Verständnis der Dynamik der erdnahen Weltraumumgebung mit Hilfe von Langzeit-Satellitenbeobachtungen N2 - The near-Earth space environment is a highly complex system comprised of several regions and particle populations hazardous to satellite operations. The trapped particles in the radiation belts and ring current can cause significant damage to satellites during space weather events, due to deep dielectric and surface charging. Closer to Earth is another important region, the ionosphere, which delays the propagation of radio signals and can adversely affect navigation and positioning. In response to fluctuations in solar and geomagnetic activity, both the inner-magnetospheric and ionospheric populations can undergo drastic and sudden changes within minutes to hours, which creates a challenge for predicting their behavior. Given the increasing reliance of our society on satellite technology, improving our understanding and modeling of these populations is a matter of paramount importance. In recent years, numerous spacecraft have been launched to study the dynamics of particle populations in the near-Earth space, transforming it into a data-rich environment. To extract valuable insights from the abundance of available observations, it is crucial to employ advanced modeling techniques, and machine learning methods are among the most powerful approaches available. This dissertation employs long-term satellite observations to analyze the processes that drive particle dynamics, and builds interdisciplinary links between space physics and machine learning by developing new state-of-the-art models of the inner-magnetospheric and ionospheric particle dynamics. The first aim of this thesis is to investigate the behavior of electrons in Earth's radiation belts and ring current. Using ~18 years of electron flux observations from the Global Positioning System (GPS), we developed the first machine learning model of hundreds-of-keV electron flux at Medium Earth Orbit (MEO) that is driven solely by solar wind and geomagnetic indices and does not require auxiliary flux measurements as inputs. We then proceeded to analyze the directional distributions of electrons, and for the first time, used Fourier sine series to fit electron pitch angle distributions (PADs) in Earth's inner magnetosphere. We performed a superposed epoch analysis of 129 geomagnetic storms during the Van Allen Probes era and demonstrated that electron PADs have a strong energy-dependent response to geomagnetic activity. Additionally, we showed that the solar wind dynamic pressure could be used as a good predictor of the PAD dynamics. Using the observed dependencies, we created the first PAD model with a continuous dependence on L, magnetic local time (MLT) and activity, and developed two techniques to reconstruct near-equatorial electron flux observations from low-PA data using this model. The second objective of this thesis is to develop a novel model of the topside ionosphere. To achieve this goal, we collected observations from five of the most widely used ionospheric missions and intercalibrated these data sets. This allowed us to use these data jointly for model development, validation, and comparison with other existing empirical models. We demonstrated, for the first time, that ion density observations by Swarm Langmuir Probes exhibit overestimation (up to ~40-50%) at low and mid-latitudes on the night side, and suggested that the influence of light ions could be a potential cause of this overestimation. To develop the topside model, we used 19 years of radio occultation (RO) electron density profiles, which were fitted with a Chapman function with a linear dependence of scale height on altitude. This approximation yields 4 parameters, namely the peak density and height of the F2-layer and the slope and intercept of the linear scale height trend, which were modeled using feedforward neural networks (NNs). The model was extensively validated against both RO and in-situ observations and was found to outperform the International Reference Ionosphere (IRI) model by up to an order of magnitude. Our analysis showed that the most substantial deviations of the IRI model from the data occur at altitudes of 100-200 km above the F2-layer peak. The developed NN-based ionospheric model reproduces the effects of various physical mechanisms observed in the topside ionosphere and provides highly accurate electron density predictions. This dissertation provides an extensive study of geospace dynamics, and the main results of this work contribute to the improvement of models of plasma populations in the near-Earth space environment. N2 - Die erdnahe Weltraumumgebung ist ein hochkomplexes System, das aus mehreren Regionen und Partikelpopulationen besteht, die für den Satellitenbetrieb gefährlich sind. Die in den Strahlungsgürteln und dem Ringstrom gefangenen Teilchen können bei Weltraumwetterereignissen aufgrund der tiefen dielektrischen und oberflächlichen Aufladung erhebliche Schäden an Satelliten verursachen. Näher an der Erde liegt eine weitere wichtige Region, die Ionosphäre, die die Ausbreitung von Funksignalen verzögert und die Navigation und Positionsbestimmung beeinträchtigen kann. Als Reaktion auf Fluktuationen der solaren und geomagnetischen Aktivität können sowohl die Populationen der inneren Magnetosphäre als auch der Ionosphäre innerhalb von Minuten bis Stunden drastische und plötzliche Veränderungen erfahren, was eine Herausforderung für die Vorhersage ihres Verhaltens darstellt. Angesichts der zunehmenden Abhängigkeit unserer Gesellschaft von der Satellitentechnologie ist ein besseres Verständnis und eine bessere Modellierung dieser Populationen von größter Bedeutung. In den letzten Jahren wurden zahlreiche Raumsonden gestartet, um die Dynamik von Partikelpopulationen im erdnahen Weltraum zu untersuchen, was diesen in eine datenreiche Umgebung verwandelt hat. Um aus der Fülle der verfügbaren Beobachtungen wertvolle Erkenntnisse zu gewinnen, ist der Einsatz fortschrittlicher Modellierungstechniken unabdingbar, und Methoden des maschinellen Lernens gehören zu den leistungsfähigsten verfügbaren Ansätzen. Diese Dissertation nutzt langfristige Satellitenbeobachtungen, um die Prozesse zu analysieren, die die Teilchendynamik antreiben, und schafft interdisziplinäre Verbindungen zwischen Weltraumphysik und maschinellem Lernen, indem sie neue hochmoderne Modelle der innermagnetosphärischen und ionosphärischen Teilchendynamik entwickelt. Das erste Ziel dieser Arbeit ist es, das Verhalten von Elektronen im Strahlungsgürtel und Ringstrom der Erde zu untersuchen. Unter Verwendung von ~18 Jahren Elektronenflussbeobachtungen des Global Positioning System (GPS) haben wir das erste maschinelle Lernmodell des Elektronenflusses im mittleren Erdorbit (MEO) entwickelt, das ausschließlich durch Sonnenwind und geomagnetische Indizes gesteuert wird und keine zusätzlichen Flussmessungen als Eingaben benötigt. Anschließend analysierten wir die Richtungsverteilungen der Elektronen und verwendeten zum ersten Mal Fourier-Sinus-Reihen, um die Elektronen-Stellwinkelverteilungen (PADs) in der inneren Magnetosphäre der Erde zu bestimmen. Wir führten eine epochenübergreifende Analyse von 129 geomagnetischen Stürmen während der Van-Allen-Sonden-Ära durch und zeigten, dass die Elektronen-PADs eine starke energieabhängige Reaktion auf die geomagnetische Aktivität haben. Außerdem konnten wir zeigen, dass der dynamische Druck des Sonnenwindes als guter Prädiktor für die PAD-Dynamik verwendet werden kann. Anhand der beobachteten Abhängigkeiten haben wir das erste PAD-Modell mit einer kontinuierlichen Abhängigkeit von L, der magnetischen Ortszeit (MLT) und der Aktivität erstellt und zwei Techniken entwickelt, um die Beobachtungen des äquatornahen Elektronenflusses aus Daten mit niedrigem Luftdruck mit Hilfe dieses Modells zu rekonstruieren. Das zweite Ziel dieser Arbeit ist die Entwicklung eines neuen Modells der Topside-Ionosphäre. Um dieses Ziel zu erreichen, haben wir Beobachtungen von fünf der meistgenutzten Ionosphärenmissionen gesammelt und diese Datensätze interkalibriert. So konnten wir diese Daten gemeinsam für die Modellentwicklung, die Validierung und den Vergleich mit anderen bestehenden empirischen Modellen nutzen. Wir haben zum ersten Mal gezeigt, dass die Ionendichtebeobachtungen von Swarm-Langmuir-Sonden in niedrigen und mittleren Breiten auf der Nachtseite eine Überschätzung (bis zu ~40-50%) aufweisen, und haben vorgeschlagen, dass der Einfluss leichter Ionen eine mögliche Ursache für diese Überschätzung sein könnte. Zur Entwicklung des Oberseitenmodells wurden 19 Jahre lang Elektronendichteprofile aus der Radio-Okkultation (RO) verwendet, die mit einer Chapman-Funktion mit einer linearen Abhängigkeit der Skalenhöhe von der Höhe angepasst wurden. Aus dieser Näherung ergeben sich 4 Parameter, nämlich die Spitzendichte und die Höhe der F2-Schicht sowie die Steigung und der Achsenabschnitt des linearen Trends der Skalenhöhe, die mit Hilfe von neuronalen Feedforward-Netzwerken (NN) modelliert wurden. Das Modell wurde sowohl anhand von RO- als auch von In-situ-Beobachtungen umfassend validiert und übertrifft das Modell der Internationalen Referenz-Ionosphäre (IRI). Unsere Analyse zeigte, dass die größten Abweichungen des IRI-Modells von den Daten in Höhen von 100-200 km über der F2-Schichtspitze auftreten. Das entwickelte NN-basierte Ionosphärenmodell reproduziert die Auswirkungen verschiedener physikalischer Mechanismen, die in der Topside-Ionosphäre beobachtet werden, und liefert sehr genaue Vorhersagen der Elektronendichte. Diese Dissertation bietet eine umfassende Untersuchung der Dynamik in der Geosphäre, und die wichtigsten Ergebnisse dieser Arbeit tragen zur Verbesserung der Modelle von Plasmapopulationen in der erdnahen Weltraumumgebung bei. KW - Ionosphere KW - radiation belts KW - ring current KW - space physics KW - empirical modeling KW - machine learning KW - gradient boosting KW - neural networks KW - Ionosphäre KW - empirische Modellierung KW - Gradient Boosting KW - maschinelles Lernen KW - neuronale Netze KW - Strahlungsgürtel KW - Ringstrom KW - Weltraumphysik Y1 - 2023 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:517-opus4-613711 ER - TY - THES A1 - Zali, Zahra T1 - Volcanic tremor analysis based on advanced signal processing concepts including music information retrieval (MIR) strategies N2 - Volcanoes are one of the Earth’s most dynamic zones and responsible for many changes in our planet. Volcano seismology aims to provide an understanding of the physical processes in volcanic systems and anticipate the style and timing of eruptions by analyzing the seismic records. Volcanic tremor signals are usually observed in the seismic records before or during volcanic eruptions. Their analysis contributes to evaluate the evolving volcanic activity and potentially predict eruptions. Years of continuous seismic monitoring now provide useful information for operational eruption forecasting. The continuously growing amount of seismic recordings, however, poses a challenge for analysis, information extraction, and interpretation, to support timely decision making during volcanic crises. Furthermore, the complexity of eruption processes and precursory activities makes the analysis challenging. A challenge in studying seismic signals of volcanic origin is the coexistence of transient signal swarms and long-lasting volcanic tremor signals. Separating transient events from volcanic tremors can, therefore, contribute to improving our understanding of the underlying physical processes. Some similar issues (data reduction, source separation, extraction, and classification) are addressed in the context of music information retrieval (MIR). The signal characteristics of acoustic and seismic recordings comprise a number of similarities. This thesis is going beyond classical signal analysis techniques usually employed in seismology by exploiting similarities of seismic and acoustic signals and building the information retrieval strategy on the expertise developed in the field of MIR. First, inspired by the idea of harmonic–percussive separation (HPS) in musical signal processing, I have developed a method to extract harmonic volcanic tremor signals and to detect transient events from seismic recordings. This provides a clean tremor signal suitable for tremor investigation along with a characteristic function suitable for earthquake detection. Second, using HPS algorithms, I have developed a noise reduction technique for seismic signals. This method is especially useful for denoising ocean bottom seismometers, which are highly contaminated by noise. The advantage of this method compared to other denoising techniques is that it doesn’t introduce distortion to the broadband earthquake waveforms, which makes it reliable for different applications in passive seismological analysis. Third, to address the challenge of extracting information from high-dimensional data and investigating the complex eruptive phases, I have developed an advanced machine learning model that results in a comprehensive signal processing scheme for volcanic tremors. Using this method seismic signatures of major eruptive phases can be automatically detected. This helps to provide a chronology of the volcanic system. Also, this model is capable to detect weak precursory volcanic tremors prior to the eruption, which could be used as an indicator of imminent eruptive activity. The extracted patterns of seismicity and their temporal variations finally provide an explanation for the transition mechanism between eruptive phases. N2 - Vulkane gehören zu den dynamischsten Zonen der Erde und sind für viele Veränderungen auf unserem Planeten verantwortlich. Die Vulkanseismologie zielt darauf ab, physikalischen Prozesse in Vulkansystemen besser zu verstehen und die Art und den Zeitpunkt von Eruptionen durch die Analyse der seismischen Aufzeichnungen vorherzusagen. Die Signale vulkanischer Tremore werden normalerweise vor oder während Vulkanausbrüchen beobachtet und müssen überwacht werden, um die vulkanische Aktivität zu bewerten. Die Untersuchung vulkanischer Tremore ist ein wichtiger Teil der Vulkanüberwachung, die darauf abzielt, Anzeichen für das Erwachen oder Wiedererwachen von Vulkanen zu erkennen und möglicherweise Ausbrüche vorherzusagen. Mehrere Dekaden kontinuierlicher seismischer Überwachung liefern nützliche Informationen für die operative Eruptionsvorhersage. Die ständig wachsende Menge an seismischen Aufzeichnungen stellt jedoch eine Herausforderung für die Analyse, Informationsextraktion und Interpretation für die zeitnahe Entscheidungsfindung während Vulkankrisen dar. Darüber hinaus erschweren die Komplexität der Eruptionsprozesse und Vorläuferaktivitäten die Analyse. Eine Herausforderung bei der Untersuchung seismischer Signale vulkanischen Ursprungs ist die Koexistenz von transienten Signalschwärmen und lang anhaltenden vulkanischen Tremoren. Die Trennung dieser beiden Signaltypen kann daher dazu beitragen, unser Verständnis der zugrunde liegenden physikalischen Prozesse zu verbessern. Einige ähnliche Probleme (Datenreduktion, Quellentrennung, Extraktion und Klassifizierung) werden im Zusammenhang mit Music Information Retrieval (MIR, dt. Etwa Musik-Informationsabruf) behandelt. Die Signaleigenschaften von akustischen und seismischen Aufzeichnungen weisen eine Reihe von Gemeinsamkeiten auf. Ich gehe über die klassischen Signalanalysetechniken hinaus, die normalerweise in der Seismologie verwendet werden, indem ich die Ähnlichkeiten von seismischen und akustischen Signalen und das Fachwissen aus dem Gebiet der MIR zur Informationsgewinnung nutze. Inspiriert von der Idee der harmonisch-perkussiven Trennung (HPS) in der musikalischen Signalverarbeitung habe ich eine Methode entwickelt, mit der harmonische vulkanische Erschütterungssignale extrahiert und transiente Ereignisse aus seismischen Aufzeichnungen erkannt werden können. Dies liefert ein sauberes Tremorsignal für die Tremoruntersuchung, sowie eine charakteristischen Funktion, die für die Erdbebenerkennung geeignet ist. Weiterhin habe ich unter Verwendung von HPS-Algorithmen eine Rauschunterdrückungstechnik für seismische Signale entwickelt. Diese kann zum Beispiel verwendet werden, um klarere Signale an Meeresbodenseismometern zu erhalten, die sonst durch zu starkes Rauschen überdeckt sind. Der Vorteil dieser Methode im Vergleich zu anderen Denoising-Techniken besteht darin, dass sie keine Verzerrung in der Breitbandantwort der Erdbebenwellen einführt, was sie für verschiedene Anwendungen in der passiven seismologischen Analyse zuverlässiger macht. Um Informationen aus hochdimensionalen Daten zu extrahieren und komplexe Eruptionsphasen zu untersuchen, habe ich ein fortschrittliches maschinelles Lernmodell entwickelt, aus dem ein umfassendes Signalverarbeitungsschema für vulkanische Erschütterungen abgeleitet werden kann. Mit dieser Methode können automatisch seismische Signaturen größerer Eruptionsphasen identifizieren werden. Dies ist nützlich, um die Chronologie eines Vulkansystems zu verstehen. Außerdem ist dieses Modell in der Lage, schwache vulkanische Vorläuferbeben zu erkennen, die als Indikator für bevorstehende Eruptionsaktivität verwendet werden könnten. Basierend auf den extrahierten Seismizitätsmustern und ihren zeitlichen Variationen liefere ich eine Erklärung für den Übergangsmechanismus zwischen verschiedenen Eruptionsphasen. KW - seismic signal processing KW - machine learning KW - volcano seismology KW - music information retrieval KW - noise reduction Y1 - 2023 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:517-opus4-610866 ER - TY - JOUR A1 - Vaid, Akhil A1 - Chan, Lili A1 - Chaudhary, Kumardeep A1 - Jaladanki, Suraj K. A1 - Paranjpe, Ishan A1 - Russak, Adam J. A1 - Kia, Arash A1 - Timsina, Prem A1 - Levin, Matthew A. A1 - He, John Cijiang A1 - Böttinger, Erwin A1 - Charney, Alexander W. A1 - Fayad, Zahi A. A1 - Coca, Steven G. A1 - Glicksberg, Benjamin S. A1 - Nadkarni, Girish N. T1 - Predictive approaches for acute dialysis requirement and death in COVID-19 JF - Clinical journal of the American Society of Nephrology : CJASN N2 - Background and objectives AKI treated with dialysis initiation is a common complication of coronavirus disease 2019 (COVID-19) among hospitalized patients. However, dialysis supplies and personnel are often limited. Design, setting, participants, & measurements Using data from adult patients hospitalized with COVID-19 from five hospitals from theMount Sinai Health System who were admitted between March 10 and December 26, 2020, we developed and validated several models (logistic regression, Least Absolute Shrinkage and Selection Operator (LASSO), random forest, and eXtreme GradientBoosting [XGBoost; with and without imputation]) for predicting treatment with dialysis or death at various time horizons (1, 3, 5, and 7 days) after hospital admission. Patients admitted to theMount Sinai Hospital were used for internal validation, whereas the other hospitals formed part of the external validation cohort. Features included demographics, comorbidities, and laboratory and vital signs within 12 hours of hospital admission. Results A total of 6093 patients (2442 in training and 3651 in external validation) were included in the final cohort. Of the different modeling approaches used, XGBoost without imputation had the highest area under the receiver operating characteristic (AUROC) curve on internal validation (range of 0.93-0.98) and area under the precisionrecall curve (AUPRC; range of 0.78-0.82) for all time points. XGBoost without imputation also had the highest test parameters on external validation (AUROC range of 0.85-0.87, and AUPRC range of 0.27-0.54) across all time windows. XGBoost without imputation outperformed all models with higher precision and recall (mean difference in AUROC of 0.04; mean difference in AUPRC of 0.15). Features of creatinine, BUN, and red cell distribution width were major drivers of the model's prediction. Conclusions An XGBoost model without imputation for prediction of a composite outcome of either death or dialysis in patients positive for COVID-19 had the best performance, as compared with standard and other machine learning models. KW - COVID-19 KW - dialysis KW - machine learning KW - prediction KW - AKI Y1 - 2021 U6 - https://doi.org/10.2215/CJN.17311120 SN - 1555-9041 SN - 1555-905X VL - 16 IS - 8 SP - 1158 EP - 1168 PB - American Society of Nephrology CY - Washington ER - TY - THES A1 - Mientus, Lukas T1 - Reflexion und Reflexivität T1 - Reflection and reflexivity BT - Befunde reflexionsbezogener Dispositionen BT - findings of reflection-related amplifiers and filters N2 - Reflexion gilt in der Lehrkräftebildung als eine Schlüsselkategorie der professionellen Entwicklung. Entsprechend wird auf vielfältige Weise die Qualität reflexionsbezogener Kompetenzen untersucht. Eine Herausforderung hierbei kann in der Annahme bestehen, von der Analyse schriftlicher Reflexionen unmittelbar auf die Reflexivität einer Person zu schließen, da Reflexion stets kontextspezifisch als Abbild reflexionsbezogener Argumentationsprozesse angesehen werden sollte und reflexionsbezogenen Dispositionen unterliegt. Auch kann die Qualität einer Reflexion auf mehreren Dimensionen bewertet werden, ohne quantifizierbare, absolute Aussagen treffen zu können. Daher wurden im Rahmen einer Physik-Videovignette N = 134 schriftliche Fremdreflexionen verfasst und kontextspezifische reflexionsbezogene Dispositionen erhoben. Expert*innen erstellten theoriegeleitet Qualitätsbewertungen zur Breite, Tiefe, Kohärenz und Spezifität eines jeden Reflexionstextes. Unter Verwendung computerbasierter Klassifikations- und Analyseverfahren wurden weitere Textmerkmale erhoben. Mittels explorativer Faktorenanalyse konnten die Faktoren Qualität, Quantität und Deskriptivität gefunden werden. Da alle konventionell eingeschätzten Qualitätsbewertungen durch einen Faktor repräsentiert wurden, konnte ein maximales Qualitätskorrelat kalkuliert werden, zu welchem jede schriftliche Fremdreflexion im Rahmen der vorliegenden Vignette eine computerbasiert bestimmbare Distanz aufweist. Diese Distanz zum maximalen Qualitätskorrelat konnte validiert werden und kann die Qualität der schriftlichen Reflexionen unabhängig von menschlichen Ressourcen quantifiziert repräsentieren. Abschließend konnte identifiziert werden, dass ausgewählte Dispositionen in unterschiedlichem Maße mit der Reflexionsqualität zusammenhängen. So konnten beispielsweise bezogen auf das Physik-Fachwissen minimale Zusammenhänge identifiziert werden, wohingegen Werthaltung sowie wahrgenommene Unterrichtsqualität eng mit der Qualität einer schriftlichen Reflexion in Verbindung stehen können. Es wird geschlussfolgert, dass reflexionsbezogene Dispositionen moderierenden Einfluss auf Reflexionen nehmen können. Es wird empfohlen bei der Erhebung von Reflexion mit dem Ziel der Kompetenzmessung ausgewählte Dispositionen mit zu erheben. Weiter verdeutlicht diese Arbeit die Möglichkeit, aussagekräftige Quantifizierungen auch in der Analyse komplexer Konstrukte vorzunehmen. Durch computerbasierte Qualitätsabschätzungen können objektive und individuelle Analysen und differenzierteres automatisiertes Feedback ermöglicht werden. N2 - Reflection is considered as a key category of professional development in teacher education. Thus, the quality of reflection-related performance has been studied in a variety of ways. To derive teacher's reflection-related personal Pedagogical Content Knowledge (PCK) from the analysis of a written reflection (reflection-related enacted PCK) seems to be challenging. The enactment of reflection-related personal PCK is context-specific and should be seen as a manifestation under the influence of Amplifiers & Filters. Also, it is difficult to make quantifiable statements of reasoning quality in a written reflection without using stage models or categorical scoring. Therefore, N = 134 (preservice) physics teachers wrote a reflection text in the context of a video vignette and answered items related to context-specific reflection-related dispositions. Experts rated the quality of each reflection text according to the breadth, depth, coherence, and specificity. Using computer-based classification and analysis, additional text features were extracted. An exploratory factor analysis was used to reduce date to the factors quality, quantity, and descriptiveness of a written reflection. Cause experts’ quality ratings were represented by just one factor, a maximum quality-correlate for the present vignette was calculated. Each written reflection was determined a distance to this maximum computer-based. This quality index was validated and can represent the quality of the written reflections in a quantified way without the need of human expertise. Finally, it could be identified that selected Amplifiers & Filters are related to the reflection quality. For example, minimal correlations could be identified with respect to physics content knowledge, whereas values and perceived teaching quality can be closely related to the quality of a written reflection. It is concluded that reflection-related Amplifiers & Filters can have a measurable influence on reflection-related enacted PCK. It is recommended to include measurements of Amplifiers & Filters in each research of reflection with the aim of measuring competence. Further, this work illustrates the possibility of meaningful quantification even in the analysis of complex constructs. Computer-based quality assessments can enable objective and individualized analyses and more differentiated automated feedback. KW - Reflexion KW - Reflexivität KW - Physikdidaktik KW - pedagogical content knowledge KW - refined consensus model KW - machine learning Y1 - 2023 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:517-opus4-610003 ER - TY - JOUR A1 - Vaid, Akhil A1 - Somani, Sulaiman A1 - Russak, Adam J. A1 - De Freitas, Jessica K. A1 - Chaudhry, Fayzan F. A1 - Paranjpe, Ishan A1 - Johnson, Kipp W. A1 - Lee, Samuel J. A1 - Miotto, Riccardo A1 - Richter, Felix A1 - Zhao, Shan A1 - Beckmann, Noam D. A1 - Naik, Nidhi A1 - Kia, Arash A1 - Timsina, Prem A1 - Lala, Anuradha A1 - Paranjpe, Manish A1 - Golden, Eddye A1 - Danieletto, Matteo A1 - Singh, Manbir A1 - Meyer, Dara A1 - O'Reilly, Paul F. A1 - Huckins, Laura A1 - Kovatch, Patricia A1 - Finkelstein, Joseph A1 - Freeman, Robert M. A1 - Argulian, Edgar A1 - Kasarskis, Andrew A1 - Percha, Bethany A1 - Aberg, Judith A. A1 - Bagiella, Emilia A1 - Horowitz, Carol R. A1 - Murphy, Barbara A1 - Nestler, Eric J. A1 - Schadt, Eric E. A1 - Cho, Judy H. A1 - Cordon-Cardo, Carlos A1 - Fuster, Valentin A1 - Charney, Dennis S. A1 - Reich, David L. A1 - Böttinger, Erwin A1 - Levin, Matthew A. A1 - Narula, Jagat A1 - Fayad, Zahi A. A1 - Just, Allan C. A1 - Charney, Alexander W. A1 - Nadkarni, Girish N. A1 - Glicksberg, Benjamin S. T1 - Machine learning to predict mortality and critical events in a cohort of patients with COVID-19 in New York City: model development and validation JF - Journal of medical internet research : international scientific journal for medical research, information and communication on the internet ; JMIR N2 - Background: COVID-19 has infected millions of people worldwide and is responsible for several hundred thousand fatalities. The COVID-19 pandemic has necessitated thoughtful resource allocation and early identification of high-risk patients. However, effective methods to meet these needs are lacking. Objective: The aims of this study were to analyze the electronic health records (EHRs) of patients who tested positive for COVID-19 and were admitted to hospitals in the Mount Sinai Health System in New York City; to develop machine learning models for making predictions about the hospital course of the patients over clinically meaningful time horizons based on patient characteristics at admission; and to assess the performance of these models at multiple hospitals and time points. Methods: We used Extreme Gradient Boosting (XGBoost) and baseline comparator models to predict in-hospital mortality and critical events at time windows of 3, 5, 7, and 10 days from admission. Our study population included harmonized EHR data from five hospitals in New York City for 4098 COVID-19-positive patients admitted from March 15 to May 22, 2020. The models were first trained on patients from a single hospital (n=1514) before or on May 1, externally validated on patients from four other hospitals (n=2201) before or on May 1, and prospectively validated on all patients after May 1 (n=383). Finally, we established model interpretability to identify and rank variables that drive model predictions. Results: Upon cross-validation, the XGBoost classifier outperformed baseline models, with an area under the receiver operating characteristic curve (AUC-ROC) for mortality of 0.89 at 3 days, 0.85 at 5 and 7 days, and 0.84 at 10 days. XGBoost also performed well for critical event prediction, with an AUC-ROC of 0.80 at 3 days, 0.79 at 5 days, 0.80 at 7 days, and 0.81 at 10 days. In external validation, XGBoost achieved an AUC-ROC of 0.88 at 3 days, 0.86 at 5 days, 0.86 at 7 days, and 0.84 at 10 days for mortality prediction. Similarly, the unimputed XGBoost model achieved an AUC-ROC of 0.78 at 3 days, 0.79 at 5 days, 0.80 at 7 days, and 0.81 at 10 days. Trends in performance on prospective validation sets were similar. At 7 days, acute kidney injury on admission, elevated LDH, tachypnea, and hyperglycemia were the strongest drivers of critical event prediction, while higher age, anion gap, and C-reactive protein were the strongest drivers of mortality prediction. Conclusions: We externally and prospectively trained and validated machine learning models for mortality and critical events for patients with COVID-19 at different time horizons. These models identified at-risk patients and uncovered underlying relationships that predicted outcomes. KW - machine learning KW - COVID-19 KW - electronic health record KW - TRIPOD KW - clinical KW - informatics KW - prediction KW - mortality KW - EHR KW - cohort KW - hospital KW - performance Y1 - 2020 U6 - https://doi.org/10.2196/24018 SN - 1439-4456 SN - 1438-8871 VL - 22 IS - 11 PB - Healthcare World CY - Richmond, Va. ER - TY - THES A1 - Seleem, Omar T1 - Towards urban pluvial flood mapping using data-driven models T1 - Kartierung städtischer Überschwemmungen mit datengesteuerten Modellen N2 - Casualties and damages from urban pluvial flooding are increasing. Triggered by short, localized, and intensive rainfall events, urban pluvial floods can occur anywhere, even in areas without a history of flooding. Urban pluvial floods have relatively small temporal and spatial scales. Although cumulative losses from urban pluvial floods are comparable, most flood risk management and mitigation strategies focus on fluvial and coastal flooding. Numerical-physical-hydrodynamic models are considered the best tool to represent the complex nature of urban pluvial floods; however, they are computationally expensive and time-consuming. These sophisticated models make large-scale analysis and operational forecasting prohibitive. Therefore, it is crucial to evaluate and benchmark the performance of other alternative methods. The findings of this cumulative thesis are represented in three research articles. The first study evaluates two topographic-based methods to map urban pluvial flooding, fill–spill–merge (FSM) and topographic wetness index (TWI), by comparing them against a sophisticated hydrodynamic model. The FSM method identifies flood-prone areas within topographic depressions while the TWI method employs maximum likelihood estimation to calibrate a TWI threshold (τ) based on inundation maps from the 2D hydrodynamic model. The results point out that the FSM method outperforms the TWI method. The study highlights then the advantage and limitations of both methods. Data-driven models provide a promising alternative to computationally expensive hydrodynamic models. However, the literature lacks benchmarking studies to evaluate the different models' performance, advantages and limitations. Model transferability in space is a crucial problem. Most studies focus on river flooding, likely due to the relative availability of flow and rain gauge records for training and validation. Furthermore, they consider these models as black boxes. The second study uses a flood inventory for the city of Berlin and 11 predictive features which potentially indicate an increased pluvial flooding hazard to map urban pluvial flood susceptibility using a convolutional neural network (CNN), an artificial neural network (ANN) and the benchmarking machine learning models random forest (RF) and support vector machine (SVM). I investigate the influence of spatial resolution on the implemented models, the models' transferability in space and the importance of the predictive features. The results show that all models perform well and the RF models are superior to the other models within and outside the training domain. The models developed using fine spatial resolution (2 and 5 m) could better identify flood-prone areas. Finally, the results point out that aspect is the most important predictive feature for the CNN models, and altitude is for the other models. While flood susceptibility maps identify flood-prone areas, they do not represent flood variables such as velocity and depth which are necessary for effective flood risk management. To address this, the third study investigates data-driven models' transferability to predict urban pluvial floodwater depth and the models' ability to enhance their predictions using transfer learning techniques. It compares the performance of RF (the best-performing model in the previous study) and CNN models using 12 predictive features and output from a hydrodynamic model. The findings in the third study suggest that while CNN models tend to generalise and smooth the target function on the training dataset, RF models suffer from overfitting. Hence, RF models are superior for predictions inside the training domains but fail outside them while CNN models could control the relative loss in performance outside the training domains. Finally, the CNN models benefit more from transfer learning techniques than RF models, boosting their performance outside training domains. In conclusion, this thesis has evaluated both topographic-based methods and data-driven models to map urban pluvial flooding. However, further studies are crucial to have methods that completely overcome the limitation of 2D hydrodynamic models. N2 - Die Zahl der Todesopfer und Schäden durch Überschwemmungen in Städten nimmt zu. Ausgelöst durch kurze, lokal begrenzte und intensive Niederschlagsereignisse können urbane pluviale Überschwemmungen überall auftreten - sogar in Gebieten, in denen es in der Vergangenheit keine Überschwemmungen gab. Urbane pluviale Überschwemmungen haben eine relativ geringe zeitliche und räumliche Ausdehnung. Obwohl die kumulativen Verluste durch urbane pluviale Überschwemmungen vergleichbar sind, konzentrieren sich die meisten Hochwasserrisikomanagement- und -minderungsstrategien auf Fluss- und Küstenüberschwemmungen. Numerisch-physikalisch-hydrodynamische Modelle gelten als das beste Instrument zur Darstellung der komplexen Natur städtischer pluvialer Überschwemmungen; sie sind jedoch rechenintensiv und zeitaufwändig. Diese anspruchsvollen Modelle machen groß angelegte Analysen und operationelle Vorhersagen unerschwinglich. Daher ist es von entscheidender Bedeutung, die Leistung anderer Methoden zu bewerten und zu vergleichen, die komplexe hydrodynamische Modelle ersetzen könnten. Die Ergebnisse dieser kumulativen Arbeit werden in drei Forschungsartikeln dargestellt. In der ersten Studie bewerte ich zwei topografiebasierte Methoden zur Kartierung von Überschwemmungen in Städten, die Fill-Spill-Merge-Methode (FSM) und den topografischen Nässeindex (TWI), indem ich sie mit einem hochentwickelten hydrodynamischen Modell vergleiche. Die FSM-Methode identifiziert überschwemmungsgefährdete Gebiete innerhalb topografischer Senken, während die TWI-Methode eine Maximum-Likelihood-Schätzung verwendet, um einen TWI-Schwellenwert (τ) auf der Grundlage von Überschwemmungskarten aus dem hydrodynamischen 2D-Modell zu kalibrieren. Die Ergebnisse zeigen, dass die FSM-Methode die TWI-Methode übertrifft. Anschließend werden die Vorteile und Grenzen beider Methoden aufgezeigt. Datengesteuerte Modelle stellen eine vielversprechende Alternative zu rechenintensiven hydrodynamischen Modellen dar. In der Literatur fehlt es jedoch an Benchmarking-Studien zur Bewertung der Leistung, Vorteile und Grenzen der verschiedenen Modelle. Die räumliche Übertragbarkeit von Modellen ist ein entscheidendes Problem. Die meisten Studien konzentrieren sich auf Flussüberschwemmungen, was wahrscheinlich auf die relative Verfügbarkeit von Abfluss- und Regenmesserdaten für Training und Validierung zurückzuführen ist. Außerdem betrachten sie diese Modelle als Black Boxes. In der zweiten Studie verwende ich ein Hochwasserinventar für die Stadt Berlin und 11 prädiktive Merkmale, die potenziell auf eine erhöhte pluviale Hochwassergefahr hinweisen, um die Anfälligkeit für pluviale Überschwemmungen in Städten zu kartieren. Dazu verwende ich ein Faltungsneuronales Netzwerk (CNN), ein Künstliches Neuronales Netzwerk (ANN) und die Benchmarking-Modelle Random Forest (RF) und Support Vector Machine (SVM). Ich untersuche den Einfluss der räumlichen Auflösung auf die implementierten Modelle, die Übertragbarkeit der Modelle im Raum und die Bedeutung der prädiktiven Merkmale. Die Ergebnisse zeigen, dass alle Modelle gut abschneiden und die RF-Modelle den anderen Modellen innerhalb und außerhalb des Trainingsbereichs überlegen sind. Die Modelle, die mit feiner räumlicher Auflösung (2 und 5 m) entwickelt wurden, konnten hochwassergefährdete Gebiete besser identifizieren. Schließlich zeigen die Ergebnisse, dass der Aspekt das wichtigste Vorhersagemerkmal für die CNN-Modelle ist, und die Höhe für die anderen Modelle. Während Hochwasseranfälligkeitskarten überschwemmungsgefährdete Gebiete identifizieren, stellen sie keine Hochwasservariablen wie Geschwindigkeit und Wassertiefe dar, die für ein effektives Hochwasserrisikomanagement notwendig sind. Um dieses Problem anzugehen, untersuche ich in der dritten Studie die Übertragbarkeit datengesteuerter Modelle auf die Vorhersage der Überschwemmungstiefe in städtischen Gebieten und die Fähigkeit der Modelle, ihre Vorhersagen durch Transfer-Learning-Techniken zu verbessern. Ich vergleiche die Leistung von RF- (das beste Modell in der vorherigen Studie) und CNN-Modellen anhand von 12 Vorhersagemerkmalen und den Ergebnissen eines hydrodynamischen Modells. Die Ergebnisse der dritten Studie deuten darauf hin, dass CNN-Modelle dazu neigen, die Zielfunktion auf dem Trainingsdatensatz zu verallgemeinern und zu glätten, während RF-Modelle unter Overfitting leiden. Daher sind RF-Modelle für Vorhersagen innerhalb der Trainingsbereiche überlegen, versagen aber außerhalb davon, während CNN-Modelle den relativen Leistungsverlust außerhalb der Trainingsdomänen kontrollieren können. Schließlich profitieren die CNN-Modelle mehr von Transfer-Learning-Techniken als RF-Modelle, was ihre Leistung außerhalb der Trainingsbereiche erhöht. Zusammenfassend lässt sich sagen, dass in dieser Arbeit sowohl topografiebasierte Methoden als auch datengesteuerte Modelle zur Kartierung von Überschwemmungen in Städten bewertet wurden. Weitere Studien sind jedoch von entscheidender Bedeutung, um Methoden zu entwickeln, die die Beschränkungen von 2D-hydrodynamischen Modellen vollständig überwinden. KW - urban pluvial flood KW - machine learning KW - deep learning KW - topography KW - tiefes Lernen KW - maschinelles Lernen KW - Topographie KW - städtische Überschwemmungen Y1 - 2023 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:517-opus4-598137 ER - TY - BOOK A1 - Weber, Benedikt T1 - Human pose estimation for decubitus prophylaxis T1 - Verwendung von Posenabschätzung zur Dekubitusprophylaxe N2 - Decubitus is one of the most relevant diseases in nursing and the most expensive to treat. It is caused by sustained pressure on tissue, so it particularly affects bed-bound patients. This work lays a foundation for pressure mattress-based decubitus prophylaxis by implementing a solution to the single-frame 2D Human Pose Estimation problem. For this, methods of Deep Learning are employed. Two approaches are examined, a coarse-to-fine Convolutional Neural Network for direct regression of joint coordinates and a U-Net for the derivation of probability distribution heatmaps. We conclude that training our models on a combined dataset of the publicly available Bodies at Rest and SLP data yields the best results. Furthermore, various preprocessing techniques are investigated, and a hyperparameter optimization is performed to discover an improved model architecture. Another finding indicates that the heatmap-based approach outperforms direct regression. This model achieves a mean per-joint position error of 9.11 cm for the Bodies at Rest data and 7.43 cm for the SLP data. We find that it generalizes well on data from mattresses other than those seen during training but has difficulties detecting the arms correctly. Additionally, we give a brief overview of the medical data annotation tool annoto we developed in the bachelor project and furthermore conclude that the Scrum framework and agile practices enhanced our development workflow. N2 - Dekubitus ist eine der relevantesten Krankheiten in der Krankenpflege und die kostspieligste in der Behandlung. Sie wird durch anhaltenden Druck auf Gewebe verursacht, betrifft also insbesondere bettlägerige Patienten. Diese Arbeit legt eine Grundlage für druckmatratzenbasierte Dekubitusprophylaxe, indem eine Lösung für das Einzelbild-2D-Posenabschätzungsproblem implementiert wird. Dafür werden Methoden des tiefen Lernens verwendet. Zwei Ansätze, basierend auf einem Gefalteten Neuronalen grob-zu-fein Netzwerk zur direkten Regression der Gelenkkoordinaten und auf einem U-Netzwerk zur Ableitung von Wahrscheinlichkeitsverteilungsbildern, werden untersucht. Wir schlussfolgern, dass das Training unserer Modelle auf einem kombinierten Datensatz, bestehend aus den frei verfügbaren Bodies at Rest und SLP Daten, die besten Ergebnisse liefert. Weiterhin werden diverse Vorverarbeitungsverfahren untersucht und eine Hyperparameteroptimierung zum Finden einer verbesserten Modellarchitektur durchgeführt. Der wahrscheinlichkeitsverteilungsbasierte Ansatz übertrifft die direkte Regression. Dieses Modell erreicht einen durchschnittlichen Pro-Gelenk-Positionsfehler von 9,11 cm auf den Bodies at Rest und von 7,43 cm auf den SLP Daten. Wir sehen, dass es gut auf Daten anderer als der im Training verwendeten Matratzen funktioniert, aber Schwierigkeiten mit der korrekten Erkennung der Arme hat. Weiterhin geben wir eine kurze Übersicht des medizinischen Datenannotationstools annoto, welches wir im Zusammenhang mit dem Bachelorprojekt entwickelt haben, und schlussfolgern außerdem, dass Scrum und agile Praktiken unseren Entwicklungsprozess verbessert haben. T3 - Technische Berichte des Hasso-Plattner-Instituts für Digital Engineering an der Universität Potsdam - 153 KW - machine learning KW - deep learning KW - convolutional neural networks KW - pose estimation KW - decubitus KW - telemedicine KW - maschinelles Lernen KW - tiefes Lernen KW - gefaltete neuronale Netze KW - Posenabschätzung KW - Dekubitus KW - Telemedizin Y1 - 2023 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:517-opus4-567196 SN - 978-3-86956-551-4 SN - 1613-5652 SN - 2191-1665 IS - 153 PB - Universitätsverlag Potsdam CY - Potsdam ER - TY - JOUR A1 - Ebers, Martin A1 - Hoch, Veronica R. S. A1 - Rosenkranz, Frank A1 - Ruschemeier, Hannah A1 - Steinrötter, Björn T1 - The European Commission’s proposal for an Artificial Intelligence Act BT - a critical assessment by members of the Robotics and AI Law Society (RAILS) JF - J : multidisciplinary scientific journal N2 - On 21 April 2021, the European Commission presented its long-awaited proposal for a Regulation “laying down harmonized rules on Artificial Intelligence”, the so-called “Artificial Intelligence Act” (AIA). This article takes a critical look at the proposed regulation. After an introduction (1), the paper analyzes the unclear preemptive effect of the AIA and EU competences (2), the scope of application (3), the prohibited uses of Artificial Intelligence (AI) (4), the provisions on high-risk AI systems (5), the obligations of providers and users (6), the requirements for AI systems with limited risks (7), the enforcement system (8), the relationship of the AIA with the existing legal framework (9), and the regulatory gaps (10). The last section draws some final conclusions (11). KW - artificial intelligence KW - machine learning KW - European Union KW - regulation KW - harmonization KW - Artificial Intelligence Act Y1 - 2021 U6 - https://doi.org/10.3390/j4040043 SN - 2571-8800 VL - 4 IS - 4 SP - 589 EP - 603 PB - MDPI CY - Basel ER - TY - BOOK A1 - Rana, Kaushik A1 - Mohapatra, Durga Prasad A1 - Sidorova, Julia A1 - Lundberg, Lars A1 - Sköld, Lars A1 - Lopes Grim, Luís Fernando A1 - Sampaio Gradvohl, André Leon A1 - Cremerius, Jonas A1 - Siegert, Simon A1 - Weltzien, Anton von A1 - Baldi, Annika A1 - Klessascheck, Finn A1 - Kalancha, Svitlana A1 - Lichtenstein, Tom A1 - Shaabani, Nuhad A1 - Meinel, Christoph A1 - Friedrich, Tobias A1 - Lenzner, Pascal A1 - Schumann, David A1 - Wiese, Ingmar A1 - Sarna, Nicole A1 - Wiese, Lena A1 - Tashkandi, Araek Sami A1 - van der Walt, Estée A1 - Eloff, Jan H. P. A1 - Schmidt, Christopher A1 - Hügle, Johannes A1 - Horschig, Siegfried A1 - Uflacker, Matthias A1 - Najafi, Pejman A1 - Sapegin, Andrey A1 - Cheng, Feng A1 - Stojanovic, Dragan A1 - Stojnev Ilić, Aleksandra A1 - Djordjevic, Igor A1 - Stojanovic, Natalija A1 - Predic, Bratislav A1 - González-Jiménez, Mario A1 - de Lara, Juan A1 - Mischkewitz, Sven A1 - Kainz, Bernhard A1 - van Hoorn, André A1 - Ferme, Vincenzo A1 - Schulz, Henning A1 - Knigge, Marlene A1 - Hecht, Sonja A1 - Prifti, Loina A1 - Krcmar, Helmut A1 - Fabian, Benjamin A1 - Ermakova, Tatiana A1 - Kelkel, Stefan A1 - Baumann, Annika A1 - Morgenstern, Laura A1 - Plauth, Max A1 - Eberhard, Felix A1 - Wolff, Felix A1 - Polze, Andreas A1 - Cech, Tim A1 - Danz, Noel A1 - Noack, Nele Sina A1 - Pirl, Lukas A1 - Beilharz, Jossekin Jakob A1 - De Oliveira, Roberto C. L. A1 - Soares, Fábio Mendes A1 - Juiz, Carlos A1 - Bermejo, Belen A1 - Mühle, Alexander A1 - Grüner, Andreas A1 - Saxena, Vageesh A1 - Gayvoronskaya, Tatiana A1 - Weyand, Christopher A1 - Krause, Mirko A1 - Frank, Markus A1 - Bischoff, Sebastian A1 - Behrens, Freya A1 - Rückin, Julius A1 - Ziegler, Adrian A1 - Vogel, Thomas A1 - Tran, Chinh A1 - Moser, Irene A1 - Grunske, Lars A1 - Szárnyas, Gábor A1 - Marton, József A1 - Maginecz, János A1 - Varró, Dániel A1 - Antal, János Benjamin ED - Meinel, Christoph ED - Polze, Andreas ED - Beins, Karsten ED - Strotmann, Rolf ED - Seibold, Ulrich ED - Rödszus, Kurt ED - Müller, Jürgen T1 - HPI Future SOC Lab – Proceedings 2018 N2 - The “HPI Future SOC Lab” is a cooperation of the Hasso Plattner Institute (HPI) and industry partners. Its mission is to enable and promote exchange and interaction between the research community and the industry partners. The HPI Future SOC Lab provides researchers with free of charge access to a complete infrastructure of state of the art hard and software. This infrastructure includes components, which might be too expensive for an ordinary research environment, such as servers with up to 64 cores and 2 TB main memory. The offerings address researchers particularly from but not limited to the areas of computer science and business information systems. Main areas of research include cloud computing, parallelization, and In-Memory technologies. This technical report presents results of research projects executed in 2018. Selected projects have presented their results on April 17th and November 14th 2017 at the Future SOC Lab Day events. N2 - Das Future SOC Lab am HPI ist eine Kooperation des Hasso-Plattner-Instituts mit verschiedenen Industriepartnern. Seine Aufgabe ist die Ermöglichung und Förderung des Austausches zwischen Forschungsgemeinschaft und Industrie. Am Lab wird interessierten Wissenschaftler:innen eine Infrastruktur von neuester Hard- und Software kostenfrei für Forschungszwecke zur Verfügung gestellt. Dazu zählen Systeme, die im normalen Hochschulbereich in der Regel nicht zu finanzieren wären, bspw. Server mit bis zu 64 Cores und 2 TB Hauptspeicher. Diese Angebote richten sich insbesondere an Wissenschaftler:innen in den Gebieten Informatik und Wirtschaftsinformatik. Einige der Schwerpunkte sind Cloud Computing, Parallelisierung und In-Memory Technologien. In diesem Technischen Bericht werden die Ergebnisse der Forschungsprojekte des Jahres 2018 vorgestellt. Ausgewählte Projekte stellten ihre Ergebnisse am 17. April und 14. November 2018 im Rahmen des Future SOC Lab Tags vor. T3 - Technische Berichte des Hasso-Plattner-Instituts für Digital Engineering an der Universität Potsdam - 151 KW - Future SOC Lab KW - research projects KW - multicore architectures KW - in-memory technology KW - cloud computing KW - machine learning KW - artifical intelligence KW - Future SOC Lab KW - Forschungsprojekte KW - Multicore Architekturen KW - In-Memory Technologie KW - Cloud Computing KW - maschinelles Lernen KW - künstliche Intelligenz Y1 - 2022 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:517-opus4-563712 SN - 978-3-86956-547-7 SN - 1613-5652 SN - 2191-1665 IS - 151 PB - Universitätsverlag Potsdam CY - Potsdam ER - TY - JOUR A1 - Schmidt, Lennart A1 - Hesse, Falk A1 - Attinger, Sabine A1 - Kumar, Rohini T1 - Challenges in applying machine learning models for hydrological inference BT - a case study for flooding events across Germany JF - Water resources research N2 - Machine learning (ML) algorithms are being increasingly used in Earth and Environmental modeling studies owing to the ever-increasing availability of diverse data sets and computational resources as well as advancement in ML algorithms. Despite advances in their predictive accuracy, the usefulness of ML algorithms for inference remains elusive. In this study, we employ two popular ML algorithms, artificial neural networks and random forest, to analyze a large data set of flood events across Germany with the goals to analyze their predictive accuracy and their usability to provide insights to hydrologic system functioning. The results of the ML algorithms are contrasted against a parametric approach based on multiple linear regression. For analysis, we employ a model-agnostic framework named Permuted Feature Importance to derive the influence of models' predictors. This allows us to compare the results of different algorithms for the first time in the context of hydrology. Our main findings are that (1) the ML models achieve higher prediction accuracy than linear regression, (2) the results reflect basic hydrological principles, but (3) further inference is hindered by the heterogeneity of results across algorithms. Thus, we conclude that the problem of equifinality as known from classical hydrological modeling also exists for ML and severely hampers its potential for inference. To account for the observed problems, we propose that when employing ML for inference, this should be made by using multiple algorithms and multiple methods, of which the latter should be embedded in a cross-validation routine. KW - machine learning KW - inference KW - floods Y1 - 2020 U6 - https://doi.org/10.1029/2019WR025924 SN - 0043-1397 SN - 1944-7973 VL - 56 IS - 5 PB - American Geophysical Union CY - Washington ER - TY - JOUR A1 - Ayzel, Georgy T1 - Deep neural networks in hydrology BT - the new generation of universal and efficient models BT - новое поколение универсальных и эффективных моделей JF - Vestnik of Saint Petersburg University. Earth Sciences N2 - For around a decade, deep learning - the sub-field of machine learning that refers to artificial neural networks comprised of many computational layers - modifies the landscape of statistical model development in many research areas, such as image classification, machine translation, and speech recognition. Geoscientific disciplines in general and the field of hydrology in particular, also do not stand aside from this movement. Recently, the proliferation of modern deep learning-based techniques and methods has been actively gaining popularity for solving a wide range of hydrological problems: modeling and forecasting of river runoff, hydrological model parameters regionalization, assessment of available water resources. identification of the main drivers of the recent change in water balance components. This growing popularity of deep neural networks is primarily due to their high universality and efficiency. The presented qualities, together with the rapidly growing amount of accumulated environmental information, as well as increasing availability of computing facilities and resources, allow us to speak about deep neural networks as a new generation of mathematical models designed to, if not to replace existing solutions, but significantly enrich the field of geophysical processes modeling. This paper provides a brief overview of the current state of the field of development and application of deep neural networks in hydrology. Also in the following study, the qualitative long-term forecast regarding the development of deep learning technology for managing the corresponding hydrological modeling challenges is provided based on the use of "Gartner Hype Curve", which in the general details describes a life cycle of modern technologies. N2 - В течение последнего десятилетия глубокое обучение - область машинного обучения, относящаяся к искусственным нейронным сетям, состоящим из множества вычислительных слоев, - изменяет ландшафт развития статистических моделей во многих областях исследований, таких как классификация изображений, машинный перевод, распознавание речи. Географические науки, а также входящая в их состав область исследования гидрологии суши, не стоят в стороне от этого движения. В последнее время применение современных технологий и методов глубокого обучения активно набирает популярность для решения широкого спектра гидрологических задач: моделирования и прогнозирования речного стока, районирования модельных параметров, оценки располагаемых водных ресурсов, идентификации факторов, влияющих на современные изменения водного режима. Такой рост популярности глубоких нейронных сетей продиктован прежде всего их высокой универсальностью и эффективностью. Представленные качества в совокупности с быстрорастущим количеством накопленной информации о состоянии окружающей среды, а также ростом доступности вычислительных средств и ресурсов, позволяют говорить о глубоких нейронных сетях как о новом поколении математических моделей, призванных если не заменить существующие решения, то значительно обогатить область моделирования геофизических процессов. В данной работе представлен краткий обзор текущего состояния области разработки и применения глубоких нейронных сетей в гидрологии. Также в работе предложен качественный долгосрочный прогноз развития технологии глубокого обучения для решения задач гидрологического моделирования на основе использования «кривой ажиотажа Гартнера», в общих чертах описывающей жизненный цикл современных технологий. T2 - Глубокие нейронные сети в гидрологии KW - deep neural networks KW - deep learning KW - machine learning KW - hydrology KW - modeling KW - глубокие нейронные сети KW - глубокое обучение KW - машинное обучение KW - гидрология KW - моделирование Y1 - 2021 U6 - https://doi.org/10.21638/spbu07.2021.101 SN - 2541-9668 SN - 2587-585X VL - 66 IS - 1 SP - 5 EP - 18 PB - Univ. Press CY - St. Petersburg ER - TY - JOUR A1 - Cope, Justin L. A1 - Baukmann, Hannes A. A1 - Klinger, Jörn E. A1 - Ravarani, Charles N. J. A1 - Böttinger, Erwin A1 - Konigorski, Stefan A1 - Schmidt, Marco F. T1 - Interaction-based feature selection algorithm outperforms polygenic risk score in predicting Parkinson’s Disease status JF - Frontiers in genetics N2 - Polygenic risk scores (PRS) aggregating results from genome-wide association studies are the state of the art in the prediction of susceptibility to complex traits or diseases, yet their predictive performance is limited for various reasons, not least of which is their failure to incorporate the effects of gene-gene interactions. Novel machine learning algorithms that use large amounts of data promise to find gene-gene interactions in order to build models with better predictive performance than PRS. Here, we present a data preprocessing step by using data-mining of contextual information to reduce the number of features, enabling machine learning algorithms to identify gene-gene interactions. We applied our approach to the Parkinson's Progression Markers Initiative (PPMI) dataset, an observational clinical study of 471 genotyped subjects (368 cases and 152 controls). With an AUC of 0.85 (95% CI = [0.72; 0.96]), the interaction-based prediction model outperforms the PRS (AUC of 0.58 (95% CI = [0.42; 0.81])). Furthermore, feature importance analysis of the model provided insights into the mechanism of Parkinson's disease. For instance, the model revealed an interaction of previously described drug target candidate genes TMEM175 and GAPDHP25. These results demonstrate that interaction-based machine learning models can improve genetic prediction models and might provide an answer to the missing heritability problem. KW - epistasis KW - machine learning KW - feature selection KW - parkinson's disease KW - PPMI (parkinson's progression markers initiative) Y1 - 2021 U6 - https://doi.org/10.3389/fgene.2021.744557 SN - 1664-8021 VL - 12 PB - Frontiers Media CY - Lausanne ER - TY - JOUR A1 - Tong, Hao A1 - Nikoloski, Zoran T1 - Machine learning approaches for crop improvement BT - leveraging phenotypic and genotypic big data JF - Journal of plant physiology : biochemistry, physiology, molecular biology and biotechnology of plants N2 - Highly efficient and accurate selection of elite genotypes can lead to dramatic shortening of the breeding cycle in major crops relevant for sustaining present demands for food, feed, and fuel. In contrast to classical approaches that emphasize the need for resource-intensive phenotyping at all stages of artificial selection, genomic selection dramatically reduces the need for phenotyping. Genomic selection relies on advances in machine learning and the availability of genotyping data to predict agronomically relevant phenotypic traits. Here we provide a systematic review of machine learning approaches applied for genomic selection of single and multiple traits in major crops in the past decade. We emphasize the need to gather data on intermediate phenotypes, e.g. metabolite, protein, and gene expression levels, along with developments of modeling techniques that can lead to further improvements of genomic selection. In addition, we provide a critical view of factors that affect genomic selection, with attention to transferability of models between different environments. Finally, we highlight the future aspects of integrating high-throughput molecular phenotypic data from omics technologies with biological networks for crop improvement. KW - genomic selection KW - genomic prediction KW - machine learning KW - multiple KW - traits KW - multi-omics KW - GxE interaction Y1 - 2020 U6 - https://doi.org/10.1016/j.jplph.2020.153354 SN - 0176-1617 SN - 1618-1328 VL - 257 PB - Elsevier CY - München ER - TY - JOUR A1 - Levy, Jessica A1 - Mussack, Dominic A1 - Brunner, Martin A1 - Keller, Ulrich A1 - Cardoso-Leite, Pedro A1 - Fischbach, Antoine T1 - Contrasting classical and machine learning approaches in the estimation of value-added scores in large-scale educational data JF - Frontiers in psychology N2 - There is no consensus on which statistical model estimates school value-added (VA) most accurately. To date, the two most common statistical models used for the calculation of VA scores are two classical methods: linear regression and multilevel models. These models have the advantage of being relatively transparent and thus understandable for most researchers and practitioners. However, these statistical models are bound to certain assumptions (e.g., linearity) that might limit their prediction accuracy. Machine learning methods, which have yielded spectacular results in numerous fields, may be a valuable alternative to these classical models. Although big data is not new in general, it is relatively new in the realm of social sciences and education. New types of data require new data analytical approaches. Such techniques have already evolved in fields with a long tradition in crunching big data (e.g., gene technology). The objective of the present paper is to competently apply these "imported" techniques to education data, more precisely VA scores, and assess when and how they can extend or replace the classical psychometrics toolbox. The different models include linear and non-linear methods and extend classical models with the most commonly used machine learning methods (i.e., random forest, neural networks, support vector machines, and boosting). We used representative data of 3,026 students in 153 schools who took part in the standardized achievement tests of the Luxembourg School Monitoring Program in grades 1 and 3. Multilevel models outperformed classical linear and polynomial regressions, as well as different machine learning models. However, it could be observed that across all schools, school VA scores from different model types correlated highly. Yet, the percentage of disagreements as compared to multilevel models was not trivial and real-life implications for individual schools may still be dramatic depending on the model type used. Implications of these results and possible ethical concerns regarding the use of machine learning methods for decision-making in education are discussed. KW - value-added modeling KW - school effectiveness KW - machine learning KW - model KW - comparison KW - longitudinal data Y1 - 2020 U6 - https://doi.org/10.3389/fpsyg.2020.02190 SN - 1664-1078 VL - 11 PB - Frontiers Research Foundation CY - Lausanne ER - TY - JOUR A1 - Steinberg, Andreas A1 - Vasyura-Bathke, Hannes A1 - Gaebler, Peter Jost A1 - Ohrnberger, Matthias A1 - Ceranna, Lars T1 - Estimation of seismic moment tensors using variational inference machine learning JF - Journal of geophysical research : Solid earth N2 - We present an approach for rapidly estimating full moment tensors of earthquakes and their parameter uncertainties based on short time windows of recorded seismic waveform data by considering deep learning of Bayesian Neural Networks (BNNs). The individual neural networks are trained on synthetic seismic waveform data and corresponding known earthquake moment-tensor parameters. A monitoring volume has been predefined to form a three-dimensional grid of locations and to train a BNN for each grid point. Variational inference on several of these networks allows us to consider several sources of error and how they affect the estimated full moment-tensor parameters and their uncertainties. In particular, we demonstrate how estimated parameter distributions are affected by uncertainties in the earthquake centroid location in space and time as well as in the assumed Earth structure model. We apply our approach as a proof of concept on seismic waveform recordings of aftershocks of the Ridgecrest 2019 earthquake with moment magnitudes ranging from Mw 2.7 to Mw 5.5. Overall, good agreement has been achieved between inferred parameter ensembles and independently estimated parameters using classical methods. Our developed approach is fast and robust, and therefore, suitable for down-stream analyses that need rapid estimates of the source mechanism for a large number of earthquakes. KW - seismology KW - machine learning KW - earthquake source KW - moment tensor KW - full KW - waveform Y1 - 2021 U6 - https://doi.org/10.1029/2021JB022685 SN - 2169-9313 SN - 2169-9356 VL - 126 IS - 10 PB - American Geophysical Union CY - Washington ER - TY - THES A1 - Chen, Junchao T1 - A self-adaptive resilient method for implementing and managing the high-reliability processing system T1 - Eine selbstadaptive belastbare Methode zum Implementieren und Verwalten von hochzuverlässigen Verarbeitungssysteme N2 - As a result of CMOS scaling, radiation-induced Single-Event Effects (SEEs) in electronic circuits became a critical reliability issue for modern Integrated Circuits (ICs) operating under harsh radiation conditions. SEEs can be triggered in combinational or sequential logic by the impact of high-energy particles, leading to destructive or non-destructive faults, resulting in data corruption or even system failure. Typically, the SEE mitigation methods are deployed statically in processing architectures based on the worst-case radiation conditions, which is most of the time unnecessary and results in a resource overhead. Moreover, the space radiation conditions are dynamically changing, especially during Solar Particle Events (SPEs). The intensity of space radiation can differ over five orders of magnitude within a few hours or days, resulting in several orders of magnitude fault probability variation in ICs during SPEs. This thesis introduces a comprehensive approach for designing a self-adaptive fault resilient multiprocessing system to overcome the static mitigation overhead issue. This work mainly addresses the following topics: (1) Design of on-chip radiation particle monitor for real-time radiation environment detection, (2) Investigation of space environment predictor, as support for solar particle events forecast, (3) Dynamic mode configuration in the resilient multiprocessing system. Therefore, according to detected and predicted in-flight space radiation conditions, the target system can be configured to use no mitigation or low-overhead mitigation during non-critical periods of time. The redundant resources can be used to improve system performance or save power. On the other hand, during increased radiation activity periods, such as SPEs, the mitigation methods can be dynamically configured appropriately depending on the real-time space radiation environment, resulting in higher system reliability. Thus, a dynamic trade-off in the target system between reliability, performance and power consumption in real-time can be achieved. All results of this work are evaluated in a highly reliable quad-core multiprocessing system that allows the self-adaptive setting of optimal radiation mitigation mechanisms during run-time. Proposed methods can serve as a basis for establishing a comprehensive self-adaptive resilient system design process. Successful implementation of the proposed design in the quad-core multiprocessor shows its application perspective also in the other designs. N2 - Infolge der CMOS-Skalierung wurden strahleninduzierte Einzelereignis-Effekte (SEEs) in elektronischen Schaltungen zu einem kritischen Zuverlässigkeitsproblem für moderne integrierte Schaltungen (ICs), die unter rauen Strahlungsbedingungen arbeiten. SEEs können in der kombinatorischen oder sequentiellen Logik durch den Aufprall hochenergetischer Teilchen ausgelöst werden, was zu destruktiven oder nicht-destruktiven Fehlern und damit zu Datenverfälschungen oder sogar Systemausfällen führt. Normalerweise werden die Methoden zur Abschwächung von SEEs statisch in Verarbeitungsarchitekturen auf der Grundlage der ungünstigsten Strahlungsbedingungen eingesetzt, was in den meisten Fällen unnötig ist und zu einem Ressourcen-Overhead führt. Darüber hinaus ändern sich die Strahlungsbedingungen im Weltraum dynamisch, insbesondere während Solar Particle Events (SPEs). Die Intensität der Weltraumstrahlung kann sich innerhalb weniger Stunden oder Tage um mehr als fünf Größenordnungen ändern, was zu einer Variation der Fehlerwahrscheinlichkeit in ICs während SPEs um mehrere Größenordnungen führt. In dieser Arbeit wird ein umfassender Ansatz für den Entwurf eines selbstanpassenden, fehlerresistenten Multiprozessorsystems vorgestellt, um das Problem des statischen Mitigation-Overheads zu überwinden. Diese Arbeit befasst sich hauptsächlich mit den folgenden Themen: (1) Entwurf eines On-Chip-Strahlungsteilchen Monitors zur Echtzeit-Erkennung von Strahlung Umgebungen, (2) Untersuchung von Weltraumumgebungsprognosen zur Unterstützung der Vorhersage von solaren Teilchen Ereignissen, (3) Konfiguration des dynamischen Modus in einem belastbaren Multiprozessorsystem. Daher kann das Zielsystem je nach den erkannten und vorhergesagten Strahlungsbedingungen während des Fluges so konfiguriert werden, dass es während unkritischer Zeiträume keine oder nur eine geringe Strahlungsminderung vornimmt. Die redundanten Ressourcen können genutzt werden, um die Systemleistung zu verbessern oder Energie zu sparen. In Zeiten erhöhter Strahlungsaktivität, wie z. B. während SPEs, können die Abschwächungsmethoden dynamisch und in Abhängigkeit von der Echtzeit-Strahlungsumgebung im Weltraum konfiguriert werden, was zu einer höheren Systemzuverlässigkeit führt. Auf diese Weise kann im Zielsystem ein dynamischer Kompromiss zwischen Zuverlässigkeit, Leistung und Stromverbrauch in Echtzeit erreicht werden. Alle Ergebnisse dieser Arbeit wurden in einem hochzuverlässigen Quad-Core-Multiprozessorsystem evaluiert, das die selbstanpassende Einstellung optimaler Strahlungsschutzmechanismen während der Laufzeit ermöglicht. Die vorgeschlagenen Methoden können als Grundlage für die Entwicklung eines umfassenden, selbstanpassenden und belastbaren Systementwurfsprozesses dienen. Die erfolgreiche Implementierung des vorgeschlagenen Entwurfs in einem Quad-Core-Multiprozessor zeigt, dass er auch für andere Entwürfe geeignet ist. KW - single event upset KW - solar particle event KW - machine learning KW - self-adaptive multiprocessing system KW - maschinelles Lernen KW - selbstanpassendes Multiprozessorsystem KW - strahleninduzierte Einzelereignis-Effekte KW - Sonnenteilchen-Ereignis Y1 - 2023 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:517-opus4-583139 ER - TY - JOUR A1 - Hecker, Pascal A1 - Steckhan, Nico A1 - Eyben, Florian A1 - Schuller, Björn Wolfgang A1 - Arnrich, Bert T1 - Voice Analysis for Neurological Disorder Recognition – A Systematic Review and Perspective on Emerging Trends JF - Frontiers in Digital Health N2 - Quantifying neurological disorders from voice is a rapidly growing field of research and holds promise for unobtrusive and large-scale disorder monitoring. The data recording setup and data analysis pipelines are both crucial aspects to effectively obtain relevant information from participants. Therefore, we performed a systematic review to provide a high-level overview of practices across various neurological disorders and highlight emerging trends. PRISMA-based literature searches were conducted through PubMed, Web of Science, and IEEE Xplore to identify publications in which original (i.e., newly recorded) datasets were collected. Disorders of interest were psychiatric as well as neurodegenerative disorders, such as bipolar disorder, depression, and stress, as well as amyotrophic lateral sclerosis amyotrophic lateral sclerosis, Alzheimer's, and Parkinson's disease, and speech impairments (aphasia, dysarthria, and dysphonia). Of the 43 retrieved studies, Parkinson's disease is represented most prominently with 19 discovered datasets. Free speech and read speech tasks are most commonly used across disorders. Besides popular feature extraction toolkits, many studies utilise custom-built feature sets. Correlations of acoustic features with psychiatric and neurodegenerative disorders are presented. In terms of analysis, statistical analysis for significance of individual features is commonly used, as well as predictive modeling approaches, especially with support vector machines and a small number of artificial neural networks. An emerging trend and recommendation for future studies is to collect data in everyday life to facilitate longitudinal data collection and to capture the behavior of participants more naturally. Another emerging trend is to record additional modalities to voice, which can potentially increase analytical performance. KW - neurological disorders KW - voice KW - speech KW - everyday life KW - multiple modalities KW - machine learning KW - disorder recognition Y1 - 2022 U6 - https://doi.org/10.3389/fdgth.2022.842301 SN - 2673-253X PB - Frontiers Media SA CY - Lausanne, Schweiz ER - TY - GEN A1 - Hecker, Pascal A1 - Steckhan, Nico A1 - Eyben, Florian A1 - Schuller, Björn Wolfgang A1 - Arnrich, Bert T1 - Voice Analysis for Neurological Disorder Recognition – A Systematic Review and Perspective on Emerging Trends T2 - Zweitveröffentlichungen der Universität Potsdam : Reihe der Digital Engineering Fakultät N2 - Quantifying neurological disorders from voice is a rapidly growing field of research and holds promise for unobtrusive and large-scale disorder monitoring. The data recording setup and data analysis pipelines are both crucial aspects to effectively obtain relevant information from participants. Therefore, we performed a systematic review to provide a high-level overview of practices across various neurological disorders and highlight emerging trends. PRISMA-based literature searches were conducted through PubMed, Web of Science, and IEEE Xplore to identify publications in which original (i.e., newly recorded) datasets were collected. Disorders of interest were psychiatric as well as neurodegenerative disorders, such as bipolar disorder, depression, and stress, as well as amyotrophic lateral sclerosis amyotrophic lateral sclerosis, Alzheimer's, and Parkinson's disease, and speech impairments (aphasia, dysarthria, and dysphonia). Of the 43 retrieved studies, Parkinson's disease is represented most prominently with 19 discovered datasets. Free speech and read speech tasks are most commonly used across disorders. Besides popular feature extraction toolkits, many studies utilise custom-built feature sets. Correlations of acoustic features with psychiatric and neurodegenerative disorders are presented. In terms of analysis, statistical analysis for significance of individual features is commonly used, as well as predictive modeling approaches, especially with support vector machines and a small number of artificial neural networks. An emerging trend and recommendation for future studies is to collect data in everyday life to facilitate longitudinal data collection and to capture the behavior of participants more naturally. Another emerging trend is to record additional modalities to voice, which can potentially increase analytical performance. T3 - Zweitveröffentlichungen der Universität Potsdam : Reihe der Digital Engineering Fakultät - 13 KW - neurological disorders KW - voice KW - speech KW - everyday life KW - multiple modalities KW - machine learning KW - disorder recognition Y1 - 2023 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:517-opus4-581019 IS - 13 ER - TY - JOUR A1 - Döllner, Jürgen Roland Friedrich T1 - Geospatial artificial intelligence BT - potentials of machine learning for 3D point clouds and geospatial digital twins JF - Journal of photogrammetry, remote sensing and geoinformation science : PFG : Photogrammetrie, Fernerkundung, Geoinformation N2 - Artificial intelligence (AI) is changing fundamentally the way how IT solutions are implemented and operated across all application domains, including the geospatial domain. This contribution outlines AI-based techniques for 3D point clouds and geospatial digital twins as generic components of geospatial AI. First, we briefly reflect on the term "AI" and outline technology developments needed to apply AI to IT solutions, seen from a software engineering perspective. Next, we characterize 3D point clouds as key category of geodata and their role for creating the basis for geospatial digital twins; we explain the feasibility of machine learning (ML) and deep learning (DL) approaches for 3D point clouds. In particular, we argue that 3D point clouds can be seen as a corpus with similar properties as natural language corpora and formulate a "Naturalness Hypothesis" for 3D point clouds. In the main part, we introduce a workflow for interpreting 3D point clouds based on ML/DL approaches that derive domain-specific and application-specific semantics for 3D point clouds without having to create explicit spatial 3D models or explicit rule sets. Finally, examples are shown how ML/DL enables us to efficiently build and maintain base data for geospatial digital twins such as virtual 3D city models, indoor models, or building information models. N2 - Georäumliche Künstliche Intelligenz: Potentiale des Maschinellen Lernens für 3D-Punktwolken und georäumliche digitale Zwillinge. Künstliche Intelligenz (KI) verändert grundlegend die Art und Weise, wie IT-Lösungen in allen Anwendungsbereichen, einschließlich dem Geoinformationsbereich, implementiert und betrieben werden. In diesem Beitrag stellen wir KI-basierte Techniken für 3D-Punktwolken als einen Baustein der georäumlichen KI vor. Zunächst werden kurz der Begriff "KI” und die technologischen Entwicklungen skizziert, die für die Anwendung von KI auf IT-Lösungen aus der Sicht der Softwaretechnik erforderlich sind. Als nächstes charakterisieren wir 3D-Punktwolken als Schlüsselkategorie von Geodaten und ihre Rolle für den Aufbau von räumlichen digitalen Zwillingen; wir erläutern die Machbarkeit der Ansätze für Maschinelles Lernen (ML) und Deep Learning (DL) in Bezug auf 3D-Punktwolken. Insbesondere argumentieren wir, dass 3D-Punktwolken als Korpus mit ähnlichen Eigenschaften wie natürlichsprachliche Korpusse gesehen werden können und formulieren eine "Natürlichkeitshypothese” für 3D-Punktwolken. Im Hauptteil stellen wir einen Workflow zur Interpretation von 3D-Punktwolken auf der Grundlage von ML/DL-Ansätzen vor, die eine domänenspezifische und anwendungsspezifische Semantik für 3D-Punktwolken ableiten, ohne explizite räumliche 3D-Modelle oder explizite Regelsätze erstellen zu müssen. Abschließend wird an Beispielen gezeigt, wie ML/DL es ermöglichen, Basisdaten für räumliche digitale Zwillinge, wie z.B. für virtuelle 3D-Stadtmodelle, Innenraummodelle oder Gebäudeinformationsmodelle, effizient aufzubauen und zu pflegen. KW - geospatial artificial intelligence KW - machine learning KW - deep learning KW - 3D KW - point clouds KW - geospatial digital twins KW - 3D city models Y1 - 2020 U6 - https://doi.org/10.1007/s41064-020-00102-3 SN - 2512-2789 SN - 2512-2819 VL - 88 IS - 1 SP - 15 EP - 24 PB - Springer International Publishing CY - Cham ER - TY - JOUR A1 - Ceulemans, Ruben A1 - Guill, Christian A1 - Gaedke, Ursula T1 - Top predators govern multitrophic diversity effects in tritrophic food webs JF - Ecology : a publication of the Ecological Society of America N2 - It is well known that functional diversity strongly affects ecosystem functioning. However, even in rather simple model communities consisting of only two or, at best, three trophic levels, the relationship between multitrophic functional diversity and ecosystem functioning appears difficult to generalize, because of its high contextuality. In this study, we considered several differently structured tritrophic food webs, in which the amount of functional diversity was varied independently on each trophic level. To achieve generalizable results, largely independent of parametrization, we examined the outcomes of 128,000 parameter combinations sampled from ecologically plausible intervals, with each tested for 200 randomly sampled initial conditions. Analysis of our data was done by training a random forest model. This method enables the identification of complex patterns in the data through partial dependence graphs, and the comparison of the relative influence of model parameters, including the degree of diversity, on food-web properties. We found that bottom-up and top-down effects cascade simultaneously throughout the food web, intimately linking the effects of functional diversity of any trophic level to the amount of diversity of other trophic levels, which may explain the difficulty in unifying results from previous studies. Strikingly, only with high diversity throughout the whole food web, different interactions synergize to ensure efficient exploitation of the available nutrients and efficient biomass transfer to higher trophic levels, ultimately leading to a high biomass and production on the top level. The temporal variation of biomass showed a more complex pattern with increasing multitrophic diversity: while the system initially became less variable, eventually the temporal variation rose again because of the increasingly complex dynamical patterns. Importantly, top predator diversity and food-web parameters affecting the top trophic level were of highest importance to determine the biomass and temporal variability of any trophic level. Overall, our study reveals that the mechanisms by which diversity influences ecosystem functioning are affected by every part of the food web, hampering the extrapolation of insights from simple monotrophic or bitrophic systems to complex natural food webs. KW - food-web efficiency KW - functional diversity KW - machine learning KW - nutrient KW - exploitation KW - production KW - random forest KW - temporal variability KW - top KW - predator KW - trait diversity Y1 - 2021 U6 - https://doi.org/10.1002/ecy.3379 SN - 0012-9658 SN - 1939-9170 VL - 102 IS - 7 PB - Wiley CY - Hoboken ER - TY - JOUR A1 - Smirnov, Artem A1 - Berrendorf, Max A1 - Shprits, Yuri Y. A1 - Kronberg, Elena A. A1 - Allison, Hayley J. A1 - Aseev, Nikita A1 - Zhelavskaya, Irina A1 - Morley, Steven K. A1 - Reeves, Geoffrey D. A1 - Carver, Matthew R. A1 - Effenberger, Frederic T1 - Medium energy electron flux in earth's outer radiation belt (MERLIN) BT - a Machine learning model JF - Space weather : the international journal of research and applications N2 - The radiation belts of the Earth, filled with energetic electrons, comprise complex and dynamic systems that pose a significant threat to satellite operation. While various models of electron flux both for low and relativistic energies have been developed, the behavior of medium energy (120-600 keV) electrons, especially in the MEO region, remains poorly quantified. At these energies, electrons are driven by both convective and diffusive transport, and their prediction usually requires sophisticated 4D modeling codes. In this paper, we present an alternative approach using the Light Gradient Boosting (LightGBM) machine learning algorithm. The Medium Energy electRon fLux In Earth's outer radiatioN belt (MERLIN) model takes as input the satellite position, a combination of geomagnetic indices and solar wind parameters including the time history of velocity, and does not use persistence. MERLIN is trained on >15 years of the GPS electron flux data and tested on more than 1.5 years of measurements. Tenfold cross validation yields that the model predicts the MEO radiation environment well, both in terms of dynamics and amplitudes o f flux. Evaluation on the test set shows high correlation between the predicted and observed electron flux (0.8) and low values of absolute error. The MERLIN model can have wide space weather applications, providing information for the scientific community in the form of radiation belts reconstructions, as well as industry for satellite mission design, nowcast of the MEO environment, and surface charging analysis. KW - machine learning KW - radiation belts KW - electron flux KW - empirical modeling KW - magnetosphere KW - electrons Y1 - 2020 U6 - https://doi.org/10.1029/2020SW002532 SN - 1542-7390 VL - 18 IS - 11 PB - American geophysical union, AGU CY - Washington ER - TY - THES A1 - Esfahani, Reza Dokht Dolatabadi T1 - Time-dependent monitoring of near-surface and ground motion modelling: developing new data processing approaches based on Music Information Retrieval (MIR) strategies N2 - Seismology, like many scientific fields, e.g., music information retrieval and speech signal pro- cessing, is experiencing exponential growth in the amount of data acquired by modern seismo- logical networks. In this thesis, I take advantage of the opportunities offered by "big data" and by the methods developed in the areas of music information retrieval and machine learning to predict better the ground motion generated by earthquakes and to study the properties of the surface layers of the Earth. In order to better predict seismic ground motions, I propose two approaches based on unsupervised deep learning methods, an autoencoder network and Generative Adversarial Networks. The autoencoder technique explores a massive amount of ground motion data, evaluates the required parameters, and generates synthetic ground motion data in the Fourier amplitude spectra (FAS) domain. This method is tested on two synthetic datasets and one real dataset. The application on the real dataset shows that the substantial information contained within the FAS data can be encoded to a four to the five-dimensional manifold. Consequently, only a few independent parameters are required for efficient ground motion prediction. I also propose a method based on Conditional Generative Adversarial Networks (CGAN) for simulating ground motion records in the time-frequency and time domains. CGAN generates the time-frequency domains based on the parameters: magnitude, distance, and shear wave velocities to 30 m depth (VS30). After generating the amplitude of the time-frequency domains using the CGAN model, instead of classical conventional methods that assume the amplitude spectra with a random phase spectrum, the phase of the time-frequency domains is recovered by minimizing the observed and reconstructed spectrograms. In the second part of this dissertation, I propose two methods for the monitoring and characterization of near-surface materials and site effect analyses. I implement an autocorrelation function and an interferometry method to monitor the velocity changes of near-surface materials resulting from the Kumamoto earthquake sequence (Japan, 2016). The observed seismic velocity changes during the strong shaking are due to the non-linear response of the near-surface materials. The results show that the velocity changes lasted for about two months after the Kumamoto mainshock. Furthermore, I used the velocity changes to evaluate the in-situ strain-stress relationship. I also propose a method for assessing the site proxy "VS30" using non-invasive analysis. In the proposed method, a dispersion curve of surface waves is inverted to estimate the shear wave velocity of the subsurface. This method is based on the Dix-like linear operators, which relate the shear wave velocity to the phase velocity. The proposed method is fast, efficient, and stable. All of the methods presented in this work can be used for processing "big data" in seismology and for the analysis of weak and strong ground motion data, to predict ground shaking, and to analyze site responses by considering potential time dependencies and nonlinearities. KW - ground motion modeling KW - machine learning KW - near-surface monitoring KW - imaging KW - generative model KW - surface wave Y1 - 2022 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:517-opus4-567671 ER - TY - JOUR A1 - Wulff, Peter A1 - Buschhüter, David A1 - Westphal, Andrea A1 - Nowak, Anna A1 - Becker, Lisa A1 - Robalino, Hugo A1 - Stede, Manfred A1 - Borowski, Andreas T1 - Computer-based classification of preservice physics teachers’ written reflections JF - Journal of science education and technology N2 - Reflecting in written form on one's teaching enactments has been considered a facilitator for teachers' professional growth in university-based preservice teacher education. Writing a structured reflection can be facilitated through external feedback. However, researchers noted that feedback in preservice teacher education often relies on holistic, rather than more content-based, analytic feedback because educators oftentimes lack resources (e.g., time) to provide more analytic feedback. To overcome this impediment to feedback for written reflection, advances in computer technology can be of use. Hence, this study sought to utilize techniques of natural language processing and machine learning to train a computer-based classifier that classifies preservice physics teachers' written reflections on their teaching enactments in a German university teacher education program. To do so, a reflection model was adapted to physics education. It was then tested to what extent the computer-based classifier could accurately classify the elements of the reflection model in segments of preservice physics teachers' written reflections. Multinomial logistic regression using word count as a predictor was found to yield acceptable average human-computer agreement (F1-score on held-out test dataset of 0.56) so that it might fuel further development towards an automated feedback tool that supplements existing holistic feedback for written reflections with data-based, analytic feedback. KW - reflection KW - teacher professional development KW - hatural language KW - processing KW - machine learning Y1 - 2020 U6 - https://doi.org/10.1007/s10956-020-09865-1 SN - 1059-0145 SN - 1573-1839 VL - 30 IS - 1 SP - 1 EP - 15 PB - Springer CY - Dordrecht ER - TY - CHAP A1 - Panzer, Marcel A1 - Bender, Benedict A1 - Gronau, Norbert T1 - Deep reinforcement learning in production planning and control BT - A systematic literature review T2 - Proceedings of the Conference on Production Systems and Logistics N2 - Increasingly fast development cycles and individualized products pose major challenges for today's smart production systems in times of industry 4.0. The systems must be flexible and continuously adapt to changing conditions while still guaranteeing high throughputs and robustness against external disruptions. Deep rein- forcement learning (RL) algorithms, which already reached impressive success with Google DeepMind's AlphaGo, are increasingly transferred to production systems to meet related requirements. Unlike supervised and unsupervised machine learning techniques, deep RL algorithms learn based on recently collected sensor- and process-data in direct interaction with the environment and are able to perform decisions in real-time. As such, deep RL algorithms seem promising given their potential to provide decision support in complex environments, as production systems, and simultaneously adapt to changing circumstances. While different use-cases for deep RL emerged, a structured overview and integration of findings on their application are missing. To address this gap, this contribution provides a systematic literature review of existing deep RL applications in the field of production planning and control as well as production logistics. From a performance perspective, it became evident that deep RL can beat heuristics significantly in their overall performance and provides superior solutions to various industrial use-cases. Nevertheless, safety and reliability concerns must be overcome before the widespread use of deep RL is possible which presumes more intensive testing of deep RL in real world applications besides the already ongoing intensive simulations. KW - deep reinforcement learning KW - machine learning KW - production planning KW - production control KW - systematic literature review Y1 - 2021 U6 - https://doi.org/10.15488/11238 SN - 2701-6277 SP - 535 EP - 545 PB - publish-Ing. CY - Hannover ER - TY - GEN A1 - Seewann, Lena A1 - Verwiebe, Roland A1 - Buder, Claudia A1 - Fritsch, Nina-Sophie T1 - “Broadcast your gender.” BT - A comparison of four text-based classification methods of German YouTube channels T2 - Zweitveröffentlichungen der Universität Potsdam : Wirtschafts- und Sozialwissenschaftliche Reihe N2 - Social media platforms provide a large array of behavioral data relevant to social scientific research. However, key information such as sociodemographic characteristics of agents are often missing. This paper aims to compare four methods of classifying social attributes from text. Specifically, we are interested in estimating the gender of German social media creators. By using the example of a random sample of 200 YouTube channels, we compare several classification methods, namely (1) a survey among university staff, (2) a name dictionary method with the World Gender Name Dictionary as a reference list, (3) an algorithmic approach using the website gender-api.com, and (4) a Multinomial Naïve Bayes (MNB) machine learning technique. These different methods identify gender attributes based on YouTube channel names and descriptions in German but are adaptable to other languages. Our contribution will evaluate the share of identifiable channels, accuracy and meaningfulness of classification, as well as limits and benefits of each approach. We aim to address methodological challenges connected to classifying gender attributes for YouTube channels as well as related to reinforcing stereotypes and ethical implications. T3 - Zweitveröffentlichungen der Universität Potsdam : Wirtschafts- und Sozialwissenschaftliche Reihe - 152 KW - text based classification methods KW - gender KW - YouTube KW - machine learning KW - authorship attribution Y1 - 2022 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:517-opus4-566287 SN - 1867-5808 IS - 152 ER - TY - JOUR A1 - Seewann, Lena A1 - Verwiebe, Roland A1 - Buder, Claudia A1 - Fritsch, Nina-Sophie T1 - “Broadcast your gender.” BT - A comparison of four text-based classification methods of German YouTube channels JF - Frontiers in Big Data N2 - Social media platforms provide a large array of behavioral data relevant to social scientific research. However, key information such as sociodemographic characteristics of agents are often missing. This paper aims to compare four methods of classifying social attributes from text. Specifically, we are interested in estimating the gender of German social media creators. By using the example of a random sample of 200 YouTube channels, we compare several classification methods, namely (1) a survey among university staff, (2) a name dictionary method with the World Gender Name Dictionary as a reference list, (3) an algorithmic approach using the website gender-api.com, and (4) a Multinomial Naïve Bayes (MNB) machine learning technique. These different methods identify gender attributes based on YouTube channel names and descriptions in German but are adaptable to other languages. Our contribution will evaluate the share of identifiable channels, accuracy and meaningfulness of classification, as well as limits and benefits of each approach. We aim to address methodological challenges connected to classifying gender attributes for YouTube channels as well as related to reinforcing stereotypes and ethical implications. KW - text based classification methods KW - gender KW - YouTube KW - machine learning KW - authorship attribution Y1 - 2022 U6 - https://doi.org/10.3389/fdata.2022.908636 SN - 2624-909X IS - 5 PB - Frontiers CY - Lausanne, Schweiz ER -