Other formats:
BibTeX
LaTeX
RIS
@inproceedings{1362555,
author = {Baisa, Vít},
address = {Brno},
booktitle = {RASLAN 2016 Recent Advances in Slavonic Natural Language Processing},
keywords = {language model; grammar agreement; verb suffix; Czech language; subject; predicate; dataset; evaluation; perplexity},
howpublished = {tištěná verze "print"},
language = {eng},
location = {Brno},
isbn = {978-80-263-1095-2},
pages = {63-67},
publisher = {Tribun EU},
title = {Czech Grammar Agreement Dataset for Evaluation of Language Models},
url = {https://nlp.fi.muni.cz/raslan/2016/paper02-Baisa.pdf},
year = {2016}
}
TY - JOUR
ID - 1362555
AU - Baisa, Vít
PY - 2016
TI - Czech Grammar Agreement Dataset for Evaluation of Language Models
PB - Tribun EU
CY - Brno
SN - 9788026310952
KW - language model
KW - grammar agreement
KW - verb suffix
KW - Czech language
KW - subject
KW - predicate
KW - dataset
KW - evaluation
KW - perplexity
UR - https://nlp.fi.muni.cz/raslan/2016/paper02-Baisa.pdf
N2 - AGREE is a dataset and task for evaluation of language models based on grammar agreement in Czech. The dataset consists of sentences with marked suffixes of past tense verbs. The task is to choose the right verb suffix which depends on gender, number and animacy of subject. It is challenging for language models because 1) Czech is morphologically rich, 2) it has relatively free word order, 3) high out-of-vocabulary (OOV) ratio, 4) predicate and subject can be far from each other, 5) subjects can be unexpressed and 6) various semantic rules may apply. The task provides a straightforward and easily reproducible way of evaluating language models on a morphologically rich language.
ER -
BAISA, Vít. Czech Grammar Agreement Dataset for Evaluation of Language Models. In \textit{RASLAN 2016 Recent Advances in Slavonic Natural Language Processing}. Brno: Tribun EU, 2016, p.~63-67. ISBN~978-80-263-1095-2.
|
