Lemmatization of Czech and Croatian Noun Clusters for Terminology Extraction
The result's identifiers
Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216224%3A14330%2F25%3A00142890" target="_blank" >RIV/00216224:14330/25:00142890 - isvavai.cz</a>
Result on the web
<a href="https://nlp.fi.muni.cz/raslan/2025/" target="_blank" >https://nlp.fi.muni.cz/raslan/2025/</a>
DOI - Digital Object Identifier
—
Alternative languages
Result language
angličtina
Original language name
Lemmatization of Czech and Croatian Noun Clusters for Terminology Extraction
Original language description
During terminology extraction, terms discovered in corpora are presented in their canonical form. Lemmatization of multi-word terms consisting of noun clusters can be ambiguous due to the lack of information on their internal structure. In this paper, we show that grammatical case alone is often not sufficient for the construction of canonical forms of noun clusters. We focus on two-noun clusters in the genitive, which are the most frequent type with ambiguous parsing. Based on corpus research, we design rules that make use of multiple morphological categories to improve the lemmatization of noun clusters found in Czech and Croatian corpora. In addition to case, we also take note of gender, animacy, and whether the noun is a proper noun. The improvements lead to more accurate and more unified forms of the terms produced during terminology extraction for these two languages in Sketch Engine.
Czech name
—
Czech description
—
Classification
Type
D - Article in proceedings
CEP classification
—
OECD FORD branch
10200 - Computer and information sciences
Result continuities
Project
<a href="/en/project/LM2023062" target="_blank" >LM2023062: Digital Research Infrastructure for Language Technologies, Arts and Humanities</a><br>
Continuities
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)
Others
Publication year
2025
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Data specific for result type
Article name in the collection
Recent Advances in Slavonic Natural Language Processing, RASLAN 2025
ISBN
9788026318583
ISSN
2336-4289
e-ISSN
—
Number of pages
8
Pages from-to
165-172
Publisher name
Tribun EU
Place of publication
Brno, Czech Republic
Event location
Kouty nad Desnou, Česká Republika
Event date
Jan 1, 2025
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
—