Preparing VerbTagGr for Modernization: Data Migration and Structural Validation of a Modern Greek Verb-Analysis Resource
Keywords:
VerbTagGr, Modern Greek, verb lemmatization, database migration, linguistic data engineeringAbstract
This paper examines the data-preparation stage in the modernization of VerbTagGr, a computational tool designed to recognize the grammatical properties of one-word verb forms in Modern Greek. The project addresses a foundational rather than algorithmic problem: how to make an established linguistic resource usable in a contemporary software environment without changing its linguistic content. VerbTagGr originated in research on statistical morphosyntactic recognition and lemmatization and was initially implemented with Java, JavaServer Pages (JSP), and MySQL. Its data resources were subsequently extended in VerbTagGr++ to cover a broader range of verb forms and to support future lemmatization. The present work focuses on the VerbDB database, especially the verb_info and ms_codeexplain tables, and documents a workflow involving MySQL configuration, database creation, schema inspection, data import, record-count verification, CSV export, and conversion to Excel workbooks for later use in Python. The equalsmscodes table was retained for structural compatibility, although it contained no records in the version examined. The resulting workbooks provide a more accessible and portable representation of the source data and establish a practical bridge between the legacy database and a planned Python implementation. The thesis reports successful completion of the conversion process and describes checks intended to preserve character encoding, field correspondence, record completeness, and the mapping between verb endings and grammatical codes. However, it does not report a benchmark of recognition accuracy or a completed Python recognizer; these remain tasks for subsequent development. The contribution is therefore best understood as a reproducible data-engineering stage that reduces technical barriers to continued research while preserving the distinction between preparing linguistic resources and validating the behavior of a complete language-processing system.
References
1. Christonakis, Μ. (2026). Upgrading the VerbTagGr language tool with a modern PC programming language [in Greek]. Diploma thesis, University of West Attica, Department of Industrial Design and Production Engineering, Athens, Greece.
2. Holton, D., Mackridge, P., & Philippaki-Warburton, E. (1999). Grammar of the Greek language [in Greek]. Patakis.
3. Kroustalli, S. (2005). VERBTAGGR web-based tool for statistical morphosyntactic recognition of one-word Modern Greek verb forms [in Greek]. Diploma thesis, National and Kapodistrian University of Athens, Department of Informatics and Telecommunications, Athens, Greece.
4. Lembessi, P. (2005). Statistical morphosyntactic recognition and lemmatization of the Modern Greek verb system. Doctoral thesis, Université Marc Bloch, Strasbourg, France.
5. Lembessi, P. (2007). VerbTagGr: Web-based tool for statistical morphosyntactic recognition of one-word Modern Greek verb forms. Paper presented at the Poland L&T Conference.
6. McKinney, W. (2010). Data structures for statistical computing in Python. In S. van der Walt & J. Millman (Eds.), Proceedings of the 9th Python in Science Conference (pp. 51–56).
7. MySQL. (2024). MySQL 8.0 Reference Manual. https://dev.mysql.com/doc/refman/8.0/en/
8. Zerva, I., Koletsi, V., & Moragianni, N. (2016). VERBTAGGR++ web environment for statistical morphosyntactic recognition of one-word Modern Greek verb forms. Bachelor’s thesis, National and Kapodistrian University of Athens, Department of Informatics and Telecommunications, Athens, Greece.
9. Klairis, C., & Babiniotis, G. (1999). Grammar of Modern Greek: Structural-functional and communicative [in Greek]. Ellinika Grammata.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Author(s)
This work is licensed under a Creative Commons Attribution 4.0 International License.
Articles published in the Journal of Advanced Multidisciplinary Studies (JAMS) are licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0), unless otherwise stated. Authors retain copyright of their work and grant JAMS the right of first publication.
