Find Academic Supervisors

Regina StoddenView profile

Research Fellow

• Text Simplification
• Online Participation
• Web Accessibility
• +6 more

Regina Stodden is a Research Fellow at the Department of Computational Linguistics, Heinrich Heine University Düsseldorf, since January 2019. She is affiliated with the NRW Research College for Online Participation (second funding phase) and works under the supervision of Prof. Dr. Marc Ziegele. Education: B.A. in Educational Science, Text Technology, and Computational Linguistics from Bielefeld University M.A. in Information Science and Language Technology from HHU Düsseldorf Her research focuses on automatic text processing , particularly text simplification for online discussions. This includes: Enabling participation for people with limited German proficiency Reducing manual workload in text analysis Exploring accessibility in Open Data portals She has contributed to tools like TS-ANNO for corpus annotation and EASSE-DE for simplification evaluation, with recent work extending to CEFR-based language proficiency assessment . Her research intersects Natural Language Processing , Machine Learning , and Usability Studies , often addressing accessibility challenges in digital participation. Scientific awards: No explicit awards mentioned. Advising and grants: Participates in the NRW Research College for Online Participation funding program and collaborates under Prof. Dr. Laura Kallmeyer's supervision. Her work involves grants related to text simplification for online participation processes.

Salar MohtajView profile

Researcher

• Transfer Learning for Natural Language Processing
• Text complexity/readability prediction
• Evaluation of Large Language Models
• +1 more

Dr. Salar Mohtaj is a senior researcher at the German Research Center for Artificial Intelligence (DFKI) and affiliated with TU Berlin's Quality and Usability Lab. He holds a PhD from TU Berlin (2024), specializing in transfer learning for harmful content detection. His work focuses on NLP applications in fake news/hate speech detection, LLM evaluation, and text complexity analysis. Education: Bachelor's: Information Technology, Shahrood University of Technology (Iran) Master's: Information Technology, Amirkabir University (Iran) PhD: TU Berlin (2024), supervised by Prof. Sebastian Möller Research Interests: Transfer learning for NLP, LLM evaluation, text complexity prediction, and detection of harmful content. He leads projects on multimodal fake news detection and semantic text similarity models. Recent Contributions: Developed the Hamtajoo plagiarism detection system, contributed to the KIALOPRIME dataset, and led EU-funded projects on AI ethics and disinformation. His work bridges NLP techniques with societal challenges like misinformation. Awards & Grants: Received Software Campus funding (2019-2024) for projects on sentiment analysis in fake news. Collaborates with Holtzbrinck Publishing Group and Deutsche Telekom. Labs & Affiliations: Speech & Language Technology group at DFKI and Quality and Usability Lab at TU Berlin. Teaches Natural Language Processing courses at TU Berlin and KI Campus.

Marco MunderlohView profile

Research Fellow

• Video Coding
• Motion Detection
• Wave Field Synthesis
• +6 more

Marco Munderloh is a research fellow at the Leibniz University Hannover under the Institute for Information Processing . His career spans advanced video coding, predictive maintenance, and photonic design, with a focus on low-bitrate aerial video compression and motion detection. Dipl.-Ing. in Computer Engineering (2004), Technical University of Ilmenau PhD (2015), Leibniz University Hannover His research interests include: Wave Field Synthesis (WFS) with patented applications in cinema sound systems Predictive maintenance using deep learning and Bayesian neural networks Inverse design for photonic integrated circuits and 3D nanostructures Region-of-interest (ROI) video coding and motion compensation Optical verification of construction materials Marco's recent publications (2025–2018) cover predictive maintenance, video compression, and photonic design. Key trends involve applying machine learning to industrial reliability, optimizing video codecs for aerial surveillance, and computational inverse design for optical devices. Scientific achievements include: Patent for WFS technology in sound reproduction Collaborations span institutions like IEEE, CIRP, EUSPEN, and journals such as APSIPA Transactions and Bautechnik . Contact: Marco.Munderloh@tnt.uni-hannover.de , Marco.Munderloh@web.de .

Thomas SchnakeView profile

Researcher

• Explainable AI
• Natural Language Processing
• XAI for Quantum Chemistry
• +3 more

Thomas Schnake is a postdoctoral researcher at the Machine Learning Lab of the Technical University of Berlin and the Berlin Institute for the Foundations of Learning and Data (BIFOLD). He holds a Ph.D. in Machine Learning from TU Berlin and prior degrees in Mathematics and Scientific Computing from Humboldt University of Berlin. His research focuses on Explainable AI (XAI), Natural Language Processing, and the mathematical foundations of machine learning. He has also gained industry experience at ebuero AG and GFaI e.V. in Berlin. Education: B.Sc. Mathematics & Philosophy, Humboldt University Berlin (2014) M.Sc. Mathematics, Humboldt University Berlin (2018) M.Sc. Scientific Computing, Technical University Berlin (2018) Ph.D. Machine Learning, Technical University Berlin (2024) Research Interests include: Explainable AI for complex domains like quantum chemistry and histopathology Graph neural network interpretability through walk-based explanations High-resolution data synthesis with minimal input Unsupervised anomaly detection in text and energy systems His recent publications (2021-2025) demonstrate contributions to XAI frameworks, graph neural network explanations, and transformer model interpretability. He is affiliated with two prominent institutions and maintains active research in interdisciplinary areas combining mathematics and machine learning.

Babak NaderiView profile

Researcher

• Subjective Quality Assessment
• Speech Quality Evaluation
• Crowdsourcing
• +3 more

Babak Naderi is a Researcher at the Quality and Usability Lab of Technical University of Berlin , where he has been employed since August 2012. He holds a Dr.-Ing. (PhD) from 2017 for his dissertation on Motivation of Workers on Microtask Crowdsourcing Platforms , a master's degree in Geodesy and Geoinformation Science from TU Berlin (2017), and a bachelor's in Software Engineering . His research focuses on subjective quality assessment , speech quality evaluation in crowdsourcing environments, and motivation theory applied to microtask platforms. He developed the Crowdwork Motivation Scale based on Self-Determination Theory and has worked on standardization efforts within the P.CROWD work program of the ITU-T Standardization Sector . His publications demonstrate expertise in crowdsourcing for speech quality assessment, statistical modeling of worker behavior, and text complexity analysis . He has contributed to IEEE and ITU-T venues while serving as a reviewer for conferences like WWW and journals including International Journal of Human-Computer Studies . Awarded BMBF training program (2013-2015) Leader of the CrowdMAQA project