TY - CHAP A1 - Klöser, Lars A1 - Kohl, Philipp A1 - Kraft, Bodo A1 - Zündorf, Albert T1 - Multi-attribute relation extraction (MARE): simplifying the application of relation extraction T2 - Proceedings of the 2nd International Conference on Deep Learning Theory and Applications - DeLTA N2 - Natural language understanding’s relation extraction makes innovative and encouraging novel business concepts possible and facilitates new digitilized decision-making processes. Current approaches allow the extraction of relations with a fixed number of entities as attributes. Extracting relations with an arbitrary amount of attributes requires complex systems and costly relation-trigger annotations to assist these systems. We introduce multi-attribute relation extraction (MARE) as an assumption-less problem formulation with two approaches, facilitating an explicit mapping from business use cases to the data annotations. Avoiding elaborated annotation constraints simplifies the application of relation extraction approaches. The evaluation compares our models to current state-of-the-art event extraction and binary relation extraction methods. Our approaches show improvement compared to these on the extraction of general multi-attribute relations. Y1 - 2021 SN - 978-989-758-526-5 U6 - http://dx.doi.org/10.5220/0010559201480156 N1 - Proceedings of the 2nd International Conference on Deep Learning Theory and Applications, DeLTA2021, July 7-9, 2021 SP - 148 EP - 156 ER - TY - JOUR A1 - Jung, Alexander A1 - Staat, Manfred T1 - Erratum to "Modeling and simulation of human induced pluripotent stem cell-derived cardiac tissue" [GAMM-Mitteilungen, (2019), 42, 4, 10.1002/gamm.201900002] JF - GAMM-Mitteilungen Y1 - 2020 U6 - http://dx.doi.org/10.1002/gamm.202000011 SN - 1522-2608 N1 - Refers to: Modeling and simulation of human induced pluripotent stem cell-derived cardiac tissue. Alexander Jung, Manfred Staat. Volume 42, Issue 4. GAMM-Mitteilungen, 2019. https://doi.org/10.1002/gamm.201900002 VL - 43 IS - 4 PB - Wiley-VCH GmbH CY - Weinheim ER - TY - GEN A1 - Jung, Alexander A1 - Müller, Wolfram A1 - Staat, Manfred T1 - Corrigendum to “Wind and fairness in ski jumping: A computer modelling analysis” [J. Biomech. 75 (2018) 147–153] T2 - Journal of Biomechanics Y1 - 2021 U6 - http://dx.doi.org/10.1016/j.jbiomech.2021.110690 SN - 0021-9290 N1 - Refers to: Alexander Jung, Wolfram Müller, Manfred Staat: Wind and fairness in ski jumping: A computer modelling analysis. Journal of Biomechanics, Volume 75. 25 June 2018. Pages 147-153. https://doi.org/10.1016/j.jbiomech.2018.05.001 VL - 128 IS - Article number: 110690 PB - Elsevier CY - Amsterdam ER - TY - BOOK A1 - Dikta, Gerhard A1 - Scheer, Marsel T1 - Bootstrap Methods: With Applications in R N2 - This book provides a compact introduction to the bootstrap method. In addition to classical results on point estimation and test theory, multivariate linear regression models and generalized linear models are covered in detail. Special attention is given to the use of bootstrap procedures to perform goodness-of-fit tests to validate model or distributional assumptions. In some cases, new methods are presented here for the first time. The text is motivated by practical examples and the implementations of the corresponding algorithms are always given directly in R in a comprehensible form. Overall, R is given great importance throughout. Each chapter includes a section of exercises and, for the more mathematically inclined readers, concludes with rigorous proofs. The intended audience is graduate students who already have a prior knowledge of probability theory and mathematical statistics. Y1 - 2021 SN - 978-3-030-73480-0 U6 - http://dx.doi.org/10.1007/978-3-030-73480-0 PB - Springer CY - Cham ER - TY - CHAP A1 - Schmidts, Oliver A1 - Kraft, Bodo A1 - Winkens, Marvin A1 - Zündorf, Albert T1 - Catalog integration of heterogeneous and volatile product data T2 - DATA 2020: Data Management Technologies and Applications N2 - The integration of frequently changing, volatile product data from different manufacturers into a single catalog is a significant challenge for small and medium-sized e-commerce companies. They rely on timely integrating product data to present them aggregated in an online shop without knowing format specifications, concept understanding of manufacturers, and data quality. Furthermore, format, concepts, and data quality may change at any time. Consequently, integrating product catalogs into a single standardized catalog is often a laborious manual task. Current strategies to streamline or automate catalog integration use techniques based on machine learning, word vectorization, or semantic similarity. However, most approaches struggle with low-quality or real-world data. We propose Attribute Label Ranking (ALR) as a recommendation engine to simplify the integration process of previously unknown, proprietary tabular format into a standardized catalog for practitioners. We evaluate ALR by focusing on the impact of different neural network architectures, language features, and semantic similarity. Additionally, we consider metrics for industrial application and present the impact of ALR in production and its limitations. Y1 - 2021 SN - 978-3-030-83013-7 U6 - http://dx.doi.org/10.1007/978-3-030-83014-4_7 N1 - International Conference on Data Management Technologies and Applications, DATA 2020, 7-9 July SP - 134 EP - 153 PB - Springer CY - Cham ER - TY - CHAP A1 - Kohl, Philipp A1 - Schmidts, Oliver A1 - Klöser, Lars A1 - Werth, Henri A1 - Kraft, Bodo A1 - Zündorf, Albert T1 - STAMP 4 NLP – an agile framework for rapid quality-driven NLP applications development T2 - Quality of Information and Communications Technology. QUATIC 2021 N2 - The progress in natural language processing (NLP) research over the last years, offers novel business opportunities for companies, as automated user interaction or improved data analysis. Building sophisticated NLP applications requires dealing with modern machine learning (ML) technologies, which impedes enterprises from establishing successful NLP projects. Our experience in applied NLP research projects shows that the continuous integration of research prototypes in production-like environments with quality assurance builds trust in the software and shows convenience and usefulness regarding the business goal. We introduce STAMP 4 NLP as an iterative and incremental process model for developing NLP applications. With STAMP 4 NLP, we merge software engineering principles with best practices from data science. Instantiating our process model allows efficiently creating prototypes by utilizing templates, conventions, and implementations, enabling developers and data scientists to focus on the business goals. Due to our iterative-incremental approach, businesses can deploy an enhanced version of the prototype to their software environment after every iteration, maximizing potential business value and trust early and avoiding the cost of successful yet never deployed experiments. KW - Machine learning KW - Process model KW - Natural language processing Y1 - 2021 SN - 978-3-030-85346-4 SN - 978-3-030-85347-1 U6 - http://dx.doi.org/10.1007/978-3-030-85347-1_12 N1 - International Conference on the Quality of Information and Communications Technology, QUATIC 2021, 8-11 September, Algarve, Portugal SP - 156 EP - 166 PB - Springer CY - Cham ER - TY - CHAP A1 - Bornheim, Tobias A1 - Grieger, Niklas A1 - Bialonski, Stephan T1 - FHAC at GermEval 2021: Identifying German toxic, engaging, and fact-claiming comments with ensemble learning T2 - Proceedings of the GermEval 2021 Workshop on the Identification of Toxic, Engaging, and Fact-Claiming Comments : 17th Conference on Natural Language Processing KONVENS 2021 Y1 - 2021 U6 - http://dx.doi.org/10.48415/2021/fhw5-x128 N1 - SP - 105 EP - 111 PB - Heinrich Heine University CY - Düsseldorf ER - TY - CHAP A1 - Engelmann, Ulrich M. A1 - Shasha, Carolyn A1 - Slabu, Ioana T1 - Magnetic nanoparticle relaxation in biomedical application: focus on simulating nanoparticle heating T2 - Magnetic nanoparticles in human health and medicine Y1 - 2021 SN - 978-1-119-75467-1 SP - 327 EP - 354 PB - Wiley-Blackwell CY - Hoboken, New Jeersey ER - TY - CHAP A1 - Olderog, M. A1 - Mohr, P. A1 - Beging, Stefan A1 - Tsoumpas, C. A1 - Ziemons, Karl T1 - Simulation study on the role of tissue-scattered events in improving sensitivity for a compact time of flight compton positron emission tomograph T2 - 2020 IEEE Nuclear Science Symposium and Medical Imaging Conference (NSS/MIC) N2 - In positron emission tomography improving time, energy and spatial detector resolutions and using Compton kinematics introduces the possibility to reconstruct a radioactivity distribution image from scatter coincidences, thereby enhancing image quality. The number of single scattered coincidences alone is in the same order of magnitude as true coincidences. In this work, a compact Compton camera module based on monolithic scintillation material is investigated as a detector ring module. The detector interactions are simulated with Monte Carlo package GATE. The scattering angle inside the tissue is derived from the energy of the scattered photon, which results in a set of possible scattering trajectories or broken line of response. The Compton kinematics collimation reduces the number of solutions. Additionally, the time of flight information helps localize the position of the annihilation. One of the questions of this investigation is related to how the energy, spatial and temporal resolutions help confine the possible annihilation volume. A comparison of currently technically feasible detector resolutions (under laboratory conditions) demonstrates the influence on this annihilation volume and shows that energy and coincidence time resolution have a significant impact. An enhancement of the latter from 400 ps to 100 ps leads to a smaller annihilation volume of around 50%, while a change of the energy resolution in the absorber layer from 12% to 4.5% results in a reduction of 60%. The inclusion of single tissue-scattered data has the potential to increase the sensitivity of a scanner by a factor of 2 to 3 times. The concept can be further optimized and extended for multiple scatter coincidences and subsequently validated by a reconstruction algorithm. Y1 - 2021 SN - 978-1-7281-7693-2 U6 - http://dx.doi.org/10.1109/NSS/MIC42677.2020.9507901 N1 - 2020 IEEE Nuclear Science Symposium and Medical Imaging Conference (NSS/MIC), 31 Oct.-7 Nov. 2020, Boston, MA, USA PB - IEEE ER - TY - JOUR A1 - Brockhaus, Moritz K. A1 - Behbahani, Mehdi A1 - Muris, Farina A1 - Jansen, Sebastian V. A1 - Schmitz- Rode, Thomas A1 - Steinseifer, Ulrich A1 - Clauser, Johanna C. T1 - In vitro thrombogenicity testing of pulsatile mechanical circulatory support systems: Design and proof-of-concept JF - Artificial Organs N2 - Thrombogenic complications are a main issue in mechanical circulatory support (MCS). There is no validated in vitro method available to quantitatively assess the thrombogenic performance of pulsatile MCS devices under realistic hemodynamic conditions. The aim of this study is to propose a method to evaluate the thrombogenic potential of new designs without the use of complex in-vivo trials. This study presents a novel in vitro method for reproducible thrombogenicity testing of pulsatile MCS systems using low molecular weight heparinized porcine blood. Blood parameters are continuously measured with full blood thromboelastometry (ROTEM; EXTEM, FIBTEM and a custom-made analysis HEPNATEM). Thrombus formation is optically observed after four hours of testing. The results of three experiments are presented each with two parallel loops. The area of thrombus formation inside the MCS device was reproducible. The implantation of a filter inside the loop catches embolizing thrombi without a measurable increase of platelet activation, allowing conclusions of the place of origin of thrombi inside the device. EXTEM and FIBTEM parameters such as clotting velocity (α) and maximum clot firmness (MCF) show a total decrease by around 6% with a characteristic kink after 180 minutes. HEPNATEM α and MCF rise within the first 180 minutes indicate a continuously increasing activation level of coagulation. After 180 minutes, the consumption of clotting factors prevails, resulting in a decrease of α and MCF. With the designed mock loop and the presented protocol we are able to identify thrombogenic hot spots inside a pulsatile pump and characterize their thrombogenic potential. Y1 - 2021 U6 - http://dx.doi.org/10.1111/aor.14046 SN - 1525-1594 VL - 45 IS - 12 SP - 1513 EP - 1521 PB - Wiley CY - Weinheim ER -