Teodor Stoev is a scientific researcher who works in the DFG project BehavE. His research interests include knowledge-based models, domain knowledge extraction, heterogeneous data, sensor data analysis, machine learning, and artificial intelligence.
Email: teodor.stoev (at) uni-greifswald.de
Phone: +49 3834 420 5505
University of Greifswald
Felix-Hausdorff-Straße 18
17489 Greifswald
Short CV
- Received the B.Sc. and M.Sc. degrees in computer science from the University of Rostock, Germany.
- From 2017 to 2020, worked as a Student Trainee and a Research Assistant at one of the biggest German direct banks—the Comdirect Bank.
- In May 2020, started working as a Researcher with the Junior Research Group, University of Rostock ‘‘Cognitive Methods for Situation-Aware Assistive Systems’’ (led by Dr.-Ing. Kristina Yordanova).
- Co-Chair of the 5th and 6th ARDUOUS Workshop on data annotation, affiliated with PerCom2021 and PerCom2022 respectively.
- In September 2022, started working at the University of Greifswald at the Institute of Data Science led by Prof. Dr.-Ing Kristina Yordanova.
Supervised Theses:
- “Entwicklung einer Pipeline zur Merkmalsextraktion und automatischen Klassifikation von Erwachsenen-Bindungs-Interviews in Deutscher Sprache” / “Developing a pipeline for feature extraction and automated Classification of Adult Attachment Interviews in German.” (Master thesis, in cooperation with Universitätsmedizin Rostock)
- “Extracting concepts and relationships for ontology generation from scraped internet forums' texts using semi‐supervised/unsupervised approaches” (Master thesis)
- “Activity recognition using the kitchen task assessment (KTA) dataset” (Master thesis)
- “Natural Language Processing für Meilensteinzuordnung im Bereich Containertransport ”/ “Natural Language Processing based milestone mapping for container shipment” (Bachelor thesis in cooperation with Ocean Insights Rostock / project44)
- “Verwendung von Texterkennungssystemen zur Informationsextraktion aus Rechnungsbelegen / “Using Text‐Recognition systems to Extract Information from Billing Receipts” (Bachelor thesis, in cooperation with vyble AG)
Publications:
- S. Suravee, T. Stoev, K. Yordanova, S. Konow (2024). Assessing Large Language Models for annotating data in Dementia-Related texts: A Comparative Study with Human Annotators. Lecture Notes in Informatics (LNI), Proceedings - Series of the Gesellschaft für Informatik (GI)
- T. Stoev, S. Suravee, K. Yordanova (2024). Variability of annotations over time. An experimental study in the dementia-related named entity recognition domain. Lecture Notes in Informatics (LNI), Proceedings - Series of the Gesellschaft für Informatik (GI)
- D. Yadav, E. Tonkin, T. Stoev, K. Yordanova (2024). A Comparative Analysis on Machine Learning Techniques for Research Metadata: the ARDUOUS Case Study. Lecture Notes in Informatics (LNI), Proceedings - Series of the Gesellschaft für Informatik (GI).
- Coping with imbalanced data in the automated detection of reminiscence from everyday life conversations of older adults. T Stoev, A Ferrario, B Demiray, M Luo, M Martin, K Yordanova. 2021. IEEE Access 9, 116540-116551
- Annotation Scheme for Named Entity Recognition and Relation Extraction Tasks in the Domain of People with Dementia. S Suravee, T Stoev, D Schindler, I Hochgraeber, C Pinkert, B Holle, Margareta Halek, Frank Krüger, Kristina Yordanova. 2022 IEEE International Conference on Pervasive Computing and Communications (PerCom)
- BehavE: Behaviour Understanding Through Automated Generation of Situation Models. T Stoev, K Yordanova. KI 2021: Advances in Artificial Intelligence., 362--369
Co-chaired Workshops:
- ARDUOUS 2021 (affiliated to the International Conference on Pervasive Computing and Communications (PerCom)).
- ARDUOUS 2022 (affiliated to the International Conference on Pervasive Computing and Communications (PerCom)).
- ARDUOUS 2023 (affiliated to the International Conference on Pervasive Computing and Communications (PerCom)).