Michal Olbrich
Main Research Interests
- Morphological Segmentation
- Multilingual Setting
- Cross-lingual Transfer
- Self Training Methods
- Language Evolution
- Computational Models of Competition
- Diachronic corpora
- Derinet
- Main author of the current version DeriNet 2.3
Projects
Derinet DeriNet
UNCE Multilingual Lens: Investigating Large Text Corpora from Different Methodological Perspectives
České neslovesné predikáty motivované substantivem a jejich syntaktické rysy
Curriculum Vitae
10/2023 – present ÚFAL - Institute of Formal and Applied Linguistics, Faculty of Mathematics and Physics, Charles University, Czech Republic
PhD thesis - Computational Models of Competition in Natural Languages, Supervisor: Zdeněk Žabokrtský
2014 – 2016 Master's degree program: Intelligent buildings, Faculty of electrical engineering, Czech technical university in Prague (ČVUT Praha)
Diploma thesis: Interrogation unit for wooden buildings monitoring
2011 – 2014 Bachelor's degree program: Cybernetics and Robotics, Faculty of electrical engineering, Czech technical university in Prague (ČVUT Praha)
Bachelor's thesis: Processing of neurophysiological data
Selected Bibliography
- Google Scholar
- ORCID: 0009-0008-2832-3211
- Scopus ID: 57193128381
- Researcher ID: JMC-1740-2023
- Olbrich, Michal, and Zdeněk Žabokrtský. "Morphological segmentation with neural networks: Performance effects of architecture, data size, and cross-lingual transfer in seven languages." International Conference on Text, Speech, and Dialogue. Cham: Springer Nature Switzerland, 2025.
- Olbrich, Michal, and Zdeněk Žabokrtský. "Improving the quality of morphological segmentation using self-training methods." DERIMO 2025 (2025): 99.
- Olbrich, Michal, and Zdenek Zabokrtsky. "Boosting the quality of morphological segmentation using transfer learning: a case study on Czech and Slovak." The 11th International Contrastive Linguistics Conference.