Naslov
Automatic Enrichment of Croatian Morphological Lexicon Using Large Corpora and Web Search

Autori
Merkler, Danijela ; Agić, Željko ; Tadić, Marko

Vrsta, podvrsta i kategorija rada
Sažeci sa skupova, pp prezentacija, znanstveni

Izvornik
Proceedings of FASSBL 2012 / - , 2012

Skup
The 8th International Conference on Formal Approaches to South Slavic and Balkan Languages (FASSBL 2012)

Mjesto i datum
Dubrovnik, Hrvatska, 19.09.2012. - 21.09.2012

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
automatic enrichment; morphological lexicon; large corpora

Sažetak
Inflectional (or morphological) lexica are considered to be language resources of high importance and frequent usage in many language processing tasks -- from basic problems such as lemmatization and morphosyntactic tagging of written text to applications in machine learning, information extraction, information retrieval and machine translation -- for highly inflectional languages such as Croatian. Being that Croatian Morphological Lexicon (HML) is frequently used both as a stand-alone application and as a module in many other systems for processing Croatian, unknown wordforms -- those undetected when matching unseen text with the current version of the HML database -- are constantly being logged and the lexicon is being updated to newer versions by inserting these new wordforms in batches. Accordingly, in this paper, we propose a generic approach to (semi-)automatic generation of new candidate lemmas for HML, their verification, assignment of inflectional patterns and finally creation and insertion of new lexicon entries to HML in a single processing pipeline.

Izvorni jezik
Engleski

Znanstvena područja
Informacijske i komunikacijske znanosti, Filologija

POVEZANOST RADA