TY - CONF
AB - This paper aims at discussing past limitations set in sentiment analysis research regarding explicit and implicit mentions of opinions. Previous studies have regularly neglected this question in favor of methodical research on standard-datasets. Furthermore, they were limited to linguistically less-diverse domains, such as commercial product reviews. We face this issue by annotating a German-language physician review dataset that contains numerous implicit, long, and complex statements that indicate aspect ratings, such as the physician’s friendliness. We discuss the nature of implicit statements and present various samples to illustrate the challenge described.
AU - Kersting, Joschka
AU - Bäumer, Frederik Simon
ED - Kersting, Joschka
ID - 31054
KW - Sentiment analysis
KW - Natural language processing
KW - Aspect phrase extraction
T2 - Proceedings of the Fourteenth International Conference on Pervasive Patterns and Applications (PATTERNS 2022): Special Track AI-DRSWA: Maturing Artificial Intelligence - Data Science for Real-World Applications
TI - Implicit Statements in Healthcare Reviews: A Challenge for Sentiment Analysis
ER -
TY - JOUR
AU - Hoyer, Britta
AU - De Jaegher, Kris
ID - 31881
JF - International Journal of Game Theory
TI - Network Disruption and the Common-Enemy Effect
ER -
TY - GEN
AB - BloKK-Beitrag für das ZeKK, 03.12.2022
AU - Lebock, Sarah
ID - 34187
TI - Blogpost "Von der Grundstimmung als philosophischer Ausgangspunkt"
ER -
TY - JOUR
AB - Far-field multi-speaker automatic speech recognition (ASR) has drawn increasing attention in recent years. Most existing methods feature a signal processing frontend and an ASR backend. In realistic scenarios, these modules are usually trained separately or progressively, which suffers from either inter-module mismatch or a complicated training process. In this paper, we propose an end-to-end multi-channel model that jointly optimizes the speech enhancement (including speech dereverberation, denoising, and separation) frontend and the ASR backend as a single system. To the best of our knowledge, this is the first work that proposes to optimize dereverberation, beamforming, and multi-speaker ASR in a fully end-to-end manner. The frontend module consists of a weighted prediction error (WPE) based submodule for dereverberation and a neural beamformer for denoising and speech separation. For the backend, we adopt a widely used end-to-end (E2E) ASR architecture. It is worth noting that the entire model is differentiable and can be optimized in a fully end-to-end manner using only the ASR criterion, without the need of parallel signal-level labels. We evaluate the proposed model on several multi-speaker benchmark datasets, and experimental results show that the fully E2E ASR model can achieve competitive performance on both noisy and reverberant conditions, with over 30% relative word error rate (WER) reduction over the single-channel baseline systems.
AU - Zhang, Wangyou
AU - Chang, Xuankai
AU - Boeddeker, Christoph
AU - Nakatani, Tomohiro
AU - Watanabe, Shinji
AU - Qian, Yanmin
ID - 33669
JF - IEEE/ACM Transactions on Audio, Speech, and Language Processing
SN - Print ISSN: 2329-9290 Electronic ISSN: 2329-9304
TI - End-to-End Dereverberation, Beamforming, and Speech Recognition in A Cocktail Party
ER -
TY - JOUR
AB - AbstractComprehensive data understanding is a key success driver for data analytics projects. Knowing the characteristics of the data helps a lot in selecting the appropriate data analysis techniques. Especially in data-driven product planning, knowledge about the data is a necessary prerequisite because data of the use phase is very heterogeneous. However, companies often do not have the necessary know-how or time to build up solid data understanding in connection with data analysis. In this paper, we develop a methodology to organize and categorize and thus understand use phase data in a way that makes it accessible to general data analytics workflows, following a design science research approach. We first present a knowledge base that lists typical use phase data from a product planning view. Second, we develop a taxonomy based on standard literature and real data objects, which covers the diversity of the data considered. The taxonomy provides 8 dimensions that support classification of use phase data and allows to capture data characteristics from a data analytics view. Finally, we combine both views by clustering the objects of the knowledge base according to the taxonomy. Each of the resulting clusters covers a typical combination of analytics relevant characteristics occurring in practice. By abstracting from the diversity of use phase data into artifacts with manageable complexity, our approach provides guidance to choose appropriate data analysis and AI techniques.
AU - Panzner, Melina
AU - von Enzberg, Sebastian
AU - Meyer, Maurice
AU - Dumitrescu, Roman
ID - 34197
JF - Journal of the Knowledge Economy
KW - Economics and Econometrics
SN - 1868-7865
TI - Characterization of Usage Data with the Help of Data Classifications
ER -
TY - JOUR
AB - Mounting sensors in disk stack separators is often a major challenge due to the operating conditions. However, a process cannot be optimally monitored without sensors. Virtual sensors can be a solution to calculate the sought parameters from measurable values. We measured the vibrations of disk stack separators and applied machine learning (ML) to detect whether the separator contains only water or whether particles are also present. We combined seven ML classification algorithms with three feature engineering strategies and evaluated our model successfully on vibration data of an experimental disk stack separator. Our experimental results demonstrate that random forest in combination with manual feature engineering using domain specific knowledge about suitable features outperforms all other models with an accuracy of 91.27 %.
AU - Merkelbach, Silke
AU - Afroze, Lameya
AU - Janssen, Nils
AU - von Enzberg, Sebastian
AU - Kühn, Arno
AU - Dumitrescu, Roman
ID - 34196
JF - Vibroengineering PROCEDIA
KW - General Medicine
SN - 2345-0533
TI - Using vibration data to classify conditions in disk stack separators
VL - 46
ER -
TY - CHAP
AU - Hobscheidt, Daniela
AU - Menzefricke, Jörn Steffen
AU - Gabriel, Stefan
AU - Kühn, Arno
AU - Dumitrescu, Roman
ID - 34195
SN - 9783658383787
T2 - Praxishandbuch Robotic Process Automation (RPA)
TI - Soziotechnische Herausforderungen bei der Einführung von RPA managen
ER -
TY - CHAP
AU - Bansmann, Michael
AU - Dumitrescu, Roman
AU - Fechtelpeter, Christian
ID - 34193
SN - 2523-3637
T2 - Gestaltung digitalisierter Arbeitswelten
TI - Transfer von Arbeit 4.0-Anwendungsszenarien
ER -
TY - CHAP
AU - Brock, Jonathan
AU - von Enzberg, Sebastian
AU - Kühn, Arno
AU - Dumitrescu, Roman
ID - 34194
SN - 9783658383787
T2 - Praxishandbuch Robotic Process Automation (RPA)
TI - Nutzung von Process Mining in RPA-Projekten
ER -
TY - JOUR
AB - Praxeologische Kompetenzansätze verstehen Kompetenz als sozial erlernt und folglich als relativ zum sozialen Kontext. Damit einher geht die Frage, wie solche praxeologisch gerahmten Kompetenzen eigentlich unabhängig von der sie hervorbringenden Praxis evaluiert werden können – und eben dadurch erst für einen breiteren Kompetenzdiskurs fruchtbar sind. Die Dokumentarische Evaluationsforschung bietet hierzu erste Anhaltspunkte, offenbart aber auch Grenzen, die mit dem Evaluationsverständnis zusammenhängen, sich jedoch in der Forschungspraxis so nicht finden lassen. Aus der Differenz zwischen Methode und Praxis dokumentarischer Evaluation lässt sich formulieren, wie eine praxeologische Evaluation gestaltet werden könnte. Dabei spielt die Formulierung von Referenzrahmen eine zentrale Rolle, welche einerseits der zu evaluierenden Praktik external sein, andererseits praktisch formuliert werden müssen, damit sie soziale Praktiken jenseits ihrer eigenen Sinnhaftigkeit evaluativ (er-)fassen können.
AU - Bloh, Thiemo
ID - 34200
IS - 02
JF - Zeitschrift für Evaluation
KW - Strategy and Management
KW - Applied Psychology
KW - Social Sciences (miscellaneous)
KW - Education
KW - Communication
KW - Statistics and Probability
SN - 1619-5515
TI - Rekonstruktive Evaluationsforschung im Kontext praxeologischer Kompetenzdiskurse. Kritische Reflexionen und konzeptionelle Überlegungen zur Dokumentarischen Evaluationsforschung
VL - 2022
ER -