Automatyczne wykrywanie nominalnych zależności referencyjnych w polskich tekstach współczesnych
Automatic detection of nominal coreference in contemporary Polish texts

Author(s): Maciej Ogrodniczuk
Subject(s): Language studies, Language and Literature Studies, Library and Information Science, Other
Published by: Wydawnictwa Uniwersytetu Warszawskiego
Keywords: reference; coreference resolution; anaphora; association; Polish language processing; National Corpus of Polish
Summary/Abstract: The book features the findings of computational linguistics research on the automatic decoding of referential relations in Polish texts. The author presents a specific definition of coreference, constructs a representative corpus of coreference using texts from the National Corpus of Polish, implements systems for detecting mentions and coreferences, and evaluates them, achieving results comparable with systems of this kind for other languages.

• Details
• Contents