TY - JOUR A1 - Mandekar, Swati A1 - Holland, Abigail A1 - Thielen, Moritz A1 - Behbahani, Mehdi A1 - Melnykowycz, Mark T1 - Advancing towards Ubiquitous EEG, Correlation of In-Ear EEG with Forehead EEG JF - Sensors N2 - Wearable EEG has gained popularity in recent years driven by promising uses outside of clinics and research. The ubiquitous application of continuous EEG requires unobtrusive form-factors that are easily acceptable by the end-users. In this progression, wearable EEG systems have been moving from full scalp to forehead and recently to the ear. The aim of this study is to demonstrate that emerging ear-EEG provides similar impedance and signal properties as established forehead EEG. EEG data using eyes-open and closed alpha paradigm were acquired from ten healthy subjects using generic earpieces fitted with three custom-made electrodes and a forehead electrode (at Fpx) after impedance analysis. Inter-subject variability in in-ear electrode impedance ranged from 20 kΩ to 25 kΩ at 10 Hz. Signal quality was comparable with an SNR of 6 for in-ear and 8 for forehead electrodes. Alpha attenuation was significant during the eyes-open condition in all in-ear electrodes, and it followed the structure of power spectral density plots of forehead electrodes, with the Pearson correlation coefficient of 0.92 between in-ear locations ELE (Left Ear Superior) and ERE (Right Ear Superior) and forehead locations, Fp1 and Fp2, respectively. The results indicate that in-ear EEG is an unobtrusive alternative in terms of impedance, signal properties and information content to established forehead EEG. KW - in-ear EEG KW - correlation KW - forehead EEG KW - impedance spectroscopy KW - biopotential electrodes Y1 - 2022 U6 - http://dx.doi.org/10.3390/s22041568 SN - 1424-8220 VL - 22 IS - 4 SP - 1 EP - 19 PB - MDPI CY - Basel ER - TY - JOUR A1 - Malinowski, Daniel A1 - Fournier, Yvan A1 - Horbach, Andreas A1 - Frick, Michael A1 - Magliani, Mirko A1 - Kalverkamp, Sebastian A1 - Hildinger, Martin A1 - Spillner, Jan A1 - Behbahani, Mehdi A1 - Hima, Flutura T1 - Computational fluid dynamics analysis of endoluminal aortic perfusion JF - Perfusion N2 - Introduction: In peripheral percutaneous (VA) extracorporeal membrane oxygenation (ECMO) procedures the femoral arteries perfusion route has inherent disadvantages regarding poor upper body perfusion due to watershed. With the advent of new long flexible cannulas an advancement of the tip up to the ascending aorta has become feasible. To investigate the impact of such long endoluminal cannulas on upper body perfusion, a Computational Fluid Dynamics (CFD) study was performed considering different support levels and three cannula positions. Methods: An idealized literature-based- and a real patient proximal aortic geometry including an endoluminal cannula were constructed. The blood flow was considered continuous. Oxygen saturation was set to 80% for the blood coming from the heart and to 100% for the blood leaving the cannula. 50% and 90% venoarterial support levels from the total blood flow rate of 6 l/min were investigated for three different positions of the cannula in the aortic arch. Results: For both geometries, the placement of the cannula in the ascending aorta led to a superior oxygenation of all aortic blood vessels except for the left coronary artery. Cannula placements at the aortic arch and descending aorta could support supra-aortic arteries, but not the coronary arteries. All positions were able to support all branches with saturated blood at 90% flow volume. Conclusions: In accordance with clinical observations CFD analysis reveals, that retrograde advancement of a long endoluminal cannula can considerably improve the oxygenation of the upper body and lead to oxygen saturation distributions similar to those of a central cannulation. KW - computational fluid dynamics analysis KW - simulation KW - endoluminal KW - aortic perfusion KW - extracorporeal membrane oxygenation Y1 - 2022 U6 - http://dx.doi.org/10.1177/02676591221099809 SN - 1477-111X VL - 0 IS - 0 SP - 1 EP - 8 PB - Sage CY - London ER - TY - JOUR A1 - Lenz, Maximilian A1 - Kahmann, Stephanie Lucina A1 - Behbahani, Mehdi A1 - Pennig, Lenhard A1 - Hackl, Michael A1 - Leschinger, Tim A1 - Müller, Lars Peter A1 - Wegmann, Kilian T1 - Influence of rotator cuff preload on fracture configuration in proximal humerus fractures: a proof of concept for fracture simulation JF - Archives of Orthopaedic and Trauma Surgery N2 - Introduction In regard of surgical training, the reproducible simulation of life-like proximal humerus fractures in human cadaveric specimens is desirable. The aim of the present study was to develop a technique that allows simulation of realistic proximal humerus fractures and to analyse the influence of rotator cuff preload on the generated lesions in regards of fracture configuration. Materials and methods Ten cadaveric specimens (6 left, 4 right) were fractured using a custom-made drop-test bench, in two groups. Five specimens were fractured without rotator cuff preload, while the other five were fractured with the tendons of the rotator cuff preloaded with 2 kg each. The humeral shaft and the shortened scapula were potted. The humerus was positioned at 90° of abduction and 10° of internal rotation to simulate a fall on the elevated arm. In two specimens of each group, the emergence of the fractures was documented with high-speed video imaging. Pre-fracture radiographs were taken to evaluate the deltoid-tuberosity index as a measure of bone density. Post-fracture X-rays and CT scans were performed to define the exact fracture configurations. Neer’s classification was used to analyse the fractures. Results In all ten cadaveric specimens life-like proximal humerus fractures were achieved. Two III-part and three IV-part fractures resulted in each group. The preloading of the rotator cuff muscles had no further influence on the fracture configuration. High-speed videos of the fracture simulation revealed identical fracture mechanisms for both groups. We observed a two-step fracture mechanism, with initial impaction of the head segment against the glenoid followed by fracturing of the head and the tuberosities and then with further impaction of the shaft against the acromion, which lead to separation of the tuberosities. Conclusion A high energetic axial impulse can reliably induce realistic proximal humerus fractures in cadaveric specimens. The preload of the rotator cuff muscles had no influence on initial fracture configuration. Therefore, fracture simulation in the proximal humerus is less elaborate. Using the presented technique, pre-fractured specimens are available for real-life surgical education. KW - Proximal humerus fracture KW - Biomechanical simulation KW - Fracture configuration KW - Fracture simulation KW - Rotator cuff Y1 - 2022 U6 - http://dx.doi.org/10.1007/s00402-022-04471-9 SN - 1434-3916 PB - Springer CY - Berlin, Heidelberg ER - TY - BOOK A1 - Laack, Walter van T1 - Greater Than the Entire Universe Y1 - 2022 SN - 978-3-936624-52-6 PB - van Laack GmbH CY - Aachen ER - TY - BOOK A1 - Laack, Walter van T1 - Aufruf zum Nachdenken: Corona und neue Kriege – Wie kann die Menschheit überleben? T1 - Call for Reflection: Corona and New Wars – How can Mankind survive? Y1 - 2022 SN - 978-3-936624-56-4 N1 - Dritter Band der Buchreihe “Vorträge & Einsichten – Lectures & Insights” und hier das zweite “Upside-Down-Buch”: Deutsche und englische Version in einem Buch. PB - van Laack GmbH CY - Aachen ER - TY - BOOK A1 - Laack, Walter van T1 - Schnittstelle Tod: Was lernen wir durch Corona über Leben und Tod? Y1 - 2022 SN - 978-3-936624-53-3 N1 - Siebter Tagungsband aus der Buchreihe “Schnittstelle Tod” PB - van Laack GmbH CY - Aachen ER - TY - JOUR A1 - Kotliar, Konstantin A1 - Ortner, Marion A1 - Conradi, Anna A1 - Hacker, Patricia A1 - Hauser, Christine A1 - Günthner, Roman A1 - Moser, Michaela A1 - Muggenthaler, Claudia A1 - Diehl-Schmid, Janine A1 - Priller, Josef A1 - Schmaderer, Christoph A1 - Grimmer, Timo T1 - Altered retinal cerebral vessel oscillation frequencies in Alzheimer's disease compatible with impaired amyloid clearance JF - Neurobiology of Aging N2 - Retinal vessels are similar to cerebral vessels in their structure and function. Moderately low oscillation frequencies of around 0.1 Hz have been reported as the driving force for paravascular drainage in gray matter in mice and are known as the frequencies of lymphatic vessels in humans. We aimed to elucidate whether retinal vessel oscillations are altered in Alzheimer's disease (AD) at the stage of dementia or mild cognitive impairment (MCI). Seventeen patients with mild-to-moderate dementia due to AD (ADD); 23 patients with MCI due to AD, and 18 cognitively healthy controls (HC) were examined using Dynamic Retinal Vessel Analyzer. Oscillatory temporal changes of retinal vessel diameters were evaluated using mathematical signal analysis. Especially at moderately low frequencies around 0.1 Hz, arterial oscillations in ADD and MCI significantly prevailed over HC oscillations and correlated with disease severity. The pronounced retinal arterial vasomotion at moderately low frequencies in the ADD and MCI groups would be compatible with the view of a compensatory upregulation of paravascular drainage in AD and strengthen the amyloid clearance hypothesis. KW - Alzheimer's disease KW - Retinal vessel analysis KW - Vasomotions KW - Pulsations KW - Mild cognitive impairment Y1 - 2022 U6 - http://dx.doi.org/10.1016/j.neurobiolaging.2022.08.012 SN - 0197-4580 VL - 120 SP - 117 EP - 127 PB - Elsevier CY - Amsterdam ER - TY - JOUR A1 - Kaulen, Lars A1 - Schwabedal, Justus T. C. A1 - Schneider, Jules A1 - Ritter, Philipp A1 - Bialonski, Stephan T1 - Advanced sleep spindle identification with neural networks JF - Scientific Reports N2 - Sleep spindles are neurophysiological phenomena that appear to be linked to memory formation and other functions of the central nervous system, and that can be observed in electroencephalographic recordings (EEG) during sleep. Manually identified spindle annotations in EEG recordings suffer from substantial intra- and inter-rater variability, even if raters have been highly trained, which reduces the reliability of spindle measures as a research and diagnostic tool. The Massive Online Data Annotation (MODA) project has recently addressed this problem by forming a consensus from multiple such rating experts, thus providing a corpus of spindle annotations of enhanced quality. Based on this dataset, we present a U-Net-type deep neural network model to automatically detect sleep spindles. Our model’s performance exceeds that of the state-of-the-art detector and of most experts in the MODA dataset. We observed improved detection accuracy in subjects of all ages, including older individuals whose spindles are particularly challenging to detect reliably. Our results underline the potential of automated methods to do repetitive cumbersome tasks with super-human performance. Y1 - 2022 U6 - http://dx.doi.org/10.1038/s41598-022-11210-y SN - 2045-2322 N1 - Corresponding author: Stephan Bialonski VL - 12 IS - Article number: 7686 SP - 1 EP - 10 PB - Springer Nature CY - London ER - TY - JOUR A1 - Herssens, Nolan A1 - Cowburn, James A1 - Albracht, Kirsten A1 - Braunstein, Bjoern A1 - Cazzola, Dario A1 - Colyer, Steffi A1 - Minetti, Alberto E. A1 - Pavei, Gaspare A1 - Rittweger, Jörn A1 - Weber, Tobias A1 - Green, David A. ED - Cattaneo, Luigi T1 - Movement in low gravity environments (MoLo) programme – the MoLo-L.O.O.P. study protocol JF - PLOS ONE / Public Library of Science N2 - Exposure to prolonged periods in microgravity is associated with deconditioning of the musculoskeletal system due to chronic changes in mechanical stimulation. Given astronauts will operate on the Lunar surface for extended periods of time, it is critical to quantify both external (e.g., ground reaction forces) and internal (e.g., joint reaction forces) loads of relevant movements performed during Lunar missions. Such knowledge is key to predict musculoskeletal deconditioning and determine appropriate exercise countermeasures associated with extended exposure to hypogravity. Y1 - 2022 U6 - http://dx.doi.org/10.1371/journal.pone.0278051 SN - 1932-6203 VL - 17 IS - 11 PB - Plos CY - San Francisco ER - TY - JOUR A1 - Gaigall, Daniel A1 - Gerstenberg, Julian A1 - Trinh, Thi Thu Ha T1 - Empirical process of concomitants for partly categorial data and applications in statistics JF - Bernoulli N2 - On the basis of independent and identically distributed bivariate random vectors, where the components are categorial and continuous variables, respectively, the related concomitants, also called induced order statistic, are considered. The main theoretical result is a functional central limit theorem for the empirical process of the concomitants in a triangular array setting. A natural application is hypothesis testing. An independence test and a two-sample test are investigated in detail. The fairly general setting enables limit results under local alternatives and bootstrap samples. For the comparison with existing tests from the literature simulation studies are conducted. The empirical results obtained confirm the theoretical findings. KW - bootstrap KW - Categorial variable KW - Concomitant KW - Empirical process KW - Independence test Y1 - 2022 U6 - http://dx.doi.org/10.3150/21-BEJ1367 SN - 1573-9759 VL - 28 IS - 2 SP - 803 EP - 829 PB - International Statistical Institute CY - Den Haag, NL ER - TY - CHAP A1 - Gaigall, Daniel T1 - On Consistent Hypothesis Testing In General Hilbert Spaces N2 - Inference on the basis of high-dimensional and functional data are two topics which are discussed frequently in the current statistical literature. A possibility to include both topics in a single approach is working on a very general space for the underlying observations, such as a separable Hilbert space. We propose a general method for consistently hypothesis testing on the basis of random variables with values in separable Hilbert spaces. We avoid concerns with the curse of dimensionality due to a projection idea. We apply well-known test statistics from nonparametric inference to the projected data and integrate over all projections from a specific set and with respect to suitable probability measures. In contrast to classical methods, which are applicable for real-valued random variables or random vectors of dimensions lower than the sample size, the tests can be applied to random vectors of dimensions larger than the sample size or even to functional and high-dimensional data. In general, resampling procedures such as bootstrap or permutation are suitable to determine critical values. The idea can be extended to the case of incomplete observations. Moreover, we develop an efficient algorithm for implementing the method. Examples are given for testing goodness-of-fit in a one-sample situation in [1] or for testing marginal homogeneity on the basis of a paired sample in [2]. Here, the test statistics in use can be seen as generalizations of the well-known Cramérvon-Mises test statistics in the one-sample and two-samples case. The treatment of other testing problems is possible as well. By using the theory of U-statistics, for instance, asymptotic null distributions of the test statistics are obtained as the sample size tends to infinity. Standard continuity assumptions ensure the asymptotic exactness of the tests under the null hypothesis and that the tests detect any alternative in the limit. Simulation studies demonstrate size and power of the tests in the finite sample case, confirm the theoretical findings, and are used for the comparison with concurring procedures. A possible application of the general approach is inference for stock market returns, also in high data frequencies. In the field of empirical finance, statistical inference of stock market prices usually takes place on the basis of related log-returns as data. In the classical models for stock prices, i.e., the exponential Lévy model, Black-Scholes model, and Merton model, properties such as independence and stationarity of the increments ensure an independent and identically structure of the data. Specific trends during certain periods of the stock price processes can cause complications in this regard. In fact, our approach can compensate those effects by the treatment of the log-returns as random vectors or even as functional data. Y1 - 2022 U6 - http://dx.doi.org/10.11159/icsta22.157 N1 - Proceedings of the 4th International Conference on Statistics: Theory and Applications (ICSTA’22) Prague, Czech Republic – July 28- 30 SP - Paper No. 157 PB - Avestia Publishing CY - Orléans, Kanada ER - TY - JOUR A1 - Engelmann, Ulrich M. A1 - Pourshahidi, Mohammad Ali A1 - Shalaby, Ahmed A1 - Krause, Hans-Joachim T1 - Probing particle size dependency of frequency mixing magnetic detection with dynamic relaxation simulation JF - Journal of Magnetism and Magnetic Materials N2 - Biomedical applications of magnetic nanoparticles (MNP) fundamentally rely on the particles’ magnetic relaxation as a response to an alternating magnetic field. The magnetic relaxation complexly depends on the interplay of MNP magnetic and physical properties with the applied field parameters. It is commonly accepted that particle core size is a major contributor to signal generation in all the above applications, however, most MNP samples comprise broad distribution spanning nm and more. Therefore, precise knowledge of the exact contribution of individual core sizes to signal generation is desired for optimal MNP design generally for each application. Specifically, we present a magnetic relaxation simulation-driven analysis of experimental frequency mixing magnetic detection (FMMD) for biosensing to quantify the contributions of individual core size fractions towards signal generation. Applying our method to two different experimental MNP systems, we found the most dominant contributions from approx. 20 nm sized particles in the two independent MNP systems. Additional comparison between freely suspended and immobilized MNP also reveals insight in the MNP microstructure, allowing to use FMMD for MNP characterization, as well as to further fine-tune its applicability in biosensing. Y1 - 2022 U6 - http://dx.doi.org/10.1016/j.jmmm.2022.169965 SN - 0304-8853 VL - 563 IS - In progress, Art. No. 169965 PB - Elsevier CY - Amsterdam ER - TY - CHAP A1 - Engelmann, Ulrich M. A1 - Baumann, Martin ED - Herbig, Nicola ED - Poppelreuter, Stefan T1 - Moderationsexpertise für QMBs – das Mindset T2 - Qualitätsmanagement im Gesundheitswesen N2 - Teamsitzungen, Arbeitsgruppentreffen, Kickoffs und Meetings – sie alle werden mit dem Ziel durchgeführt, innerhalb einer vorgegebenen Zeitspanne ein gemeinsames Arbeitsziel zu erreichen. Damit die Zielerreichung auch bei komplexeren Arbeitsaufträgen nicht vom Zufall abhängt, empfiehlt es sich, die Leitung des Ablaufs einem Moderator zu übertragen. In diesem Beitrag einer mehrteiligen Serie wird beschrieben, über welches Mindset der Moderator verfügen sollte, welche grundsätzlichen Methoden hilfreich sind und was bei der Onlinemoderation im Besonderen zu beachten ist. Y1 - 2022 SN - 978-3-8249-0714-4 SP - Kapitel 10814 PB - TÜV-Verlag CY - Köln ET - 59. Update ER - TY - BOOK A1 - Engelmann, Ulrich M. T1 - Zielführend moderieren N2 - In der Teamarbeit wird Moderation zum Erfolgsfaktor, der jedoch häufig unterschätzt wird. Ausgehend vom persönlichen Kompetenzniveau verknüpft dieses Buch Grundlagen und Methoden zu Wegen, um Ihre persönliche Entwicklung individuell zu begleiten: Neulinge finden hilfreiche Checklisten und Basistechniken für ihre ersten Moderationen, Fortgeschrittene wertvolle Praxistipps und Methoden für den Ausbau ihrer Moderationskompetenz. Profis schließlich genießen eine raffinierte Aussicht auf weniger bekannte Techniken und neue Anwendungen. Weiterführende Exkurse zum Meeting-Management und zur Online-Moderation runden den Anwendungshorizont ab. Ob in Beruf, Studium oder Ehrenamt – derart ausgestattet gelingen Ihre eigene sowie die Entwicklung Ihres Teams durch zielführende Moderationen. Y1 - 2022 SN - 9783838556895 U6 - http://dx.doi.org/10.36198/9783838556895 PB - UVK Verlag CY - Stuttgart ER - TY - JOUR A1 - Ditzhaus, Marc A1 - Gaigall, Daniel T1 - Testing marginal homogeneity in Hilbert spaces with applications to stock market returns JF - Test N2 - This paper considers a paired data framework and discusses the question of marginal homogeneity of bivariate high-dimensional or functional data. The related testing problem can be endowed into a more general setting for paired random variables taking values in a general Hilbert space. To address this problem, a Cramér–von-Mises type test statistic is applied and a bootstrap procedure is suggested to obtain critical values and finally a consistent test. The desired properties of a bootstrap test can be derived that are asymptotic exactness under the null hypothesis and consistency under alternatives. Simulations show the quality of the test in the finite sample case. A possible application is the comparison of two possibly dependent stock market returns based on functional data. The approach is demonstrated based on historical data for different stock market indices. Y1 - 2022 U6 - http://dx.doi.org/10.1007/s11749-022-00802-5 SN - 1863-8260 VL - 2022 IS - 31 SP - 749 EP - 770 PB - Springer ER - TY - JOUR A1 - Defosse, Jerome A1 - Kleinschmidt, Joris A1 - Schmutz, Axel A1 - Loop, Torsten A1 - Staat, Manfred A1 - Gatzweiler, Karl-Heinz A1 - Wappler, Frank A1 - Schieren, Mark T1 - Dental strain on maxillary incisors during tracheal intubation with double-lumen tubes and different laryngoscopy techniques - a blinded manikin study JF - Journal of Cardiothoracic and Vascular Anesthesia KW - anaesthetic complications KW - dental trauma KW - difficult airway KW - double-lumen tube intubation KW - videolaryngoscopy Y1 - 2022 U6 - http://dx.doi.org/10.1053/j.jvca.2022.02.017 SN - 1053-0770 VL - 36 IS - 8, Part B SP - 3021 EP - 3027 PB - Elsevier CY - New York, NY ER - TY - JOUR A1 - Colombo, Daniele A1 - Drira, Slah A1 - Frotscher, Ralf A1 - Staat, Manfred T1 - An element-based formulation for ES-FEM and FS-FEM models for implementation in standard solid mechanics finite element codes for 2D and 3D static analysis JF - International Journal for Numerical Methods in Engineering N2 - Edge-based and face-based smoothed finite element methods (ES-FEM and FS-FEM, respectively) are modified versions of the finite element method allowing to achieve more accurate results and to reduce sensitivity to mesh distortion, at least for linear elements. These properties make the two methods very attractive. However, their implementation in a standard finite element code is nontrivial because it requires heavy and extensive modifications to the code architecture. In this article, we present an element-based formulation of ES-FEM and FS-FEM methods allowing to implement the two methods in a standard finite element code with no modifications to its architecture. Moreover, the element-based formulation permits to easily manage any type of element, especially in 3D models where, to the best of the authors' knowledge, only tetrahedral elements are used in FS-FEM applications found in the literature. Shape functions for non-simplex 3D elements are proposed in order to apply FS-FEM to any standard finite element. KW - distorted element KW - ES-FEM KW - FS-FEM KW - non-simplex S-FEM elements KW - S-FEM Y1 - 2022 U6 - http://dx.doi.org/10.1002/nme.7126 SN - 1097-0207 VL - 124 IS - 2 SP - 402 EP - 433 PB - Wiley CY - Chichester ER - TY - JOUR A1 - Chloé, Radermacher A1 - Malyaran, Hanna A1 - Craveiro, Rogerio Bastos A1 - Peglow, Sarah A1 - Behbahani, Mehdi A1 - Pufe, Thomas A1 - Wolf, Michael A1 - Neuss, Sabine T1 - Mechanical loading on cementoblasts: a mini review JF - Osteologie N2 - Orthodontic treatments are concomitant with mechanical forces and thereby cause teeth movements. The applied forces are transmitted to the tooth root and the periodontal ligaments which is compressed on one side and tensed up on the other side. Indeed, strong forces can lead to tooth root resorption and the crown-to-tooth ratio is reduced with the potential for significant clinical impact. The cementum, which covers the tooth root, is a thin mineralized tissue of the periodontium that connects the periodontal ligament with the tooth and is build up by cementoblasts. The impact of tension and compression on these cells is investigated in several in vivo and in vitro studies demonstrating differences in protein expression and signaling pathways. In summary, osteogenic marker changes indicate that cyclic tensile forces support whereas static tension inhibits cementogenesis. Furthermore, cementogenesis experiences the same protein expression changes in static conditions as static tension, but cyclic compression leads to the exact opposite of cyclic tension. Consistent with marker expression changes, the singaling pathways of Wnt/ß-catenin and RANKL/OPG show that tissue compression leads to cementum degradation and tension forces to cementogenesis. However, the cementum, and in particular its cementoblasts, remain a research area which should be explored in more detail to understand the underlying mechanism of bone resorption and remodeling after orthodontic treatments. KW - Cementoblast KW - Compression KW - Tension KW - Mechanotransduction KW - Forces Y1 - 2022 U6 - http://dx.doi.org/10.1055/a-1826-0777 SN - 1019-1291 VL - 31 IS - 2 SP - 111 EP - 118 PB - Thieme CY - Stuttgart ER - TY - CHAP A1 - Büsgen, André A1 - Klöser, Lars A1 - Kohl, Philipp A1 - Schmidts, Oliver A1 - Kraft, Bodo A1 - Zündorf, Albert T1 - Exploratory analysis of chat-based black market profiles with natural language processing T2 - Proceedings of the 11th International Conference on Data Science, Technology and Applications N2 - Messenger apps like WhatsApp or Telegram are an integral part of daily communication. Besides the various positive effects, those services extend the operating range of criminals. Open trading groups with many thousand participants emerged on Telegram. Law enforcement agencies monitor suspicious users in such chat rooms. This research shows that text analysis, based on natural language processing, facilitates this through a meaningful domain overview and detailed investigations. We crawled a corpus from such self-proclaimed black markets and annotated five attribute types products, money, payment methods, user names, and locations. Based on each message a user sends, we extract and group these attributes to build profiles. Then, we build features to cluster the profiles. Pretrained word vectors yield better unsupervised clustering results than current state-of-the-art transformer models. The result is a semantically meaningful high-level overview of the user landscape of black market chatrooms. Additionally, the extracted structured information serves as a foundation for further data exploration, for example, the most active users or preferred payment methods. KW - Clustering KW - Natural Language Processing KW - Information Extraction KW - Profile Extraction KW - Text Mining Y1 - 2022 SN - 978-989-758-583-8 U6 - http://dx.doi.org/10.5220/0011271400003269 SN - 2184-285X SP - 83 EP - 94 ER - TY - CHAP A1 - Blaneck, Patrick Gustav A1 - Bornheim, Tobias A1 - Grieger, Niklas A1 - Bialonski, Stephan T1 - Automatic readability assessment of german sentences with transformer ensembles T2 - Proceedings of the GermEval 2022 Workshop on Text Complexity Assessment of German Text N2 - Reliable methods for automatic readability assessment have the potential to impact a variety of fields, ranging from machine translation to self-informed learning. Recently, large language models for the German language (such as GBERT and GPT-2-Wechsel) have become available, allowing to develop Deep Learning based approaches that promise to further improve automatic readability assessment. In this contribution, we studied the ability of ensembles of fine-tuned GBERT and GPT-2-Wechsel models to reliably predict the readability of German sentences. We combined these models with linguistic features and investigated the dependence of prediction performance on ensemble size and composition. Mixed ensembles of GBERT and GPT-2-Wechsel performed better than ensembles of the same size consisting of only GBERT or GPT-2-Wechsel models. Our models were evaluated in the GermEval 2022 Shared Task on Text Complexity Assessment on data of German sentences. On out-of-sample data, our best ensemble achieved a root mean squared error of 0:435. Y1 - 2022 U6 - http://dx.doi.org/10.48550/arXiv.2209.04299 N1 - Proceedings of the 18th Conference on Natural Language Processing/Konferenz zur Verarbeitung natürlicher Sprache (KONVENS 2022) 12-15 September, 2022 University of Potsdam Potsdam, Germany SP - 57 EP - 62 PB - Association for Computational Linguistics CY - Potsdam ER -