|
|
The data published in this repository were prepared at the Department of Latvian and Baltic Studies, Faculty of Humanities, University of Latvia, as part of the project Database of Latvian Morphemes and Derivational Models (DLMDM), project No. lzp-2022/1-0013, funded by the Latvian Council of Science (2023–2026). Project leader: Dr. philol. Andra Kalnača, Professor at the Department of Latvian and Baltic Studies, Faculty of Humanities, University of Latvia (andra.kalnaca).
Database authors:
Andra Kalnača, Tatjana Pakalne (Editors-in-chief)
Ieva Auziņa, Vanesa Balmane, Anita Butāne, Milan Hoplíček, Daiki Horiguchi, Laura Paula Jansone, Kristīne Levāne‑Petrova, Ilze Lokmane, Paula Miķelsone, Oskars Otomers, Paula Ozola, Inta Urbanoviča.
This repository contains four linguistic registers. Each register captures a different layer of derivational morphology data and is provided as a UTF-8 encoded, tab-separated file.
| Register | Description | File path |
|---|---|---|
| Lemma register | Lexical entries with segmentation, part of speech, morphological features, links to root hierarchies, and parent relations. | data/lemma_register.tsv |
| Root register | Roots with allomorphs and hierarchical relations, linked to lemmas containing those roots. | data/root_register.tsv |
| Affix register | Derivational and inflectional affixes with structural and functional information. | data/affix_register.tsv |
| Source register | Sources of lemmas. | data/source_register.tsv |
data/— the four registers.docs/— project overview and data model description.docs/examples/— examples of isolated word families as encoded in the registers, with visualizations.CHANGELOG.md— version history.LICENSE— usage and redistribution terms.README.md— project description and usage notes.
The dataset follows semantic versioning. The version history and the current version are stored in the CHANGELOG.md file.
If you use these registers in research, please cite the DLMDM project:
Kalnača, Andra, Pakalne, Tatjana (eds.). 2026. Database of Latvian Morphemes and Derivational Models (DLMDM). Rīga: University of Latvia. Available at: https://github.com/MorphLatLang/DLMDM
E-mail: andra.kalnaca@lu.lv, tatjana.pakalne@lu.lv