@inproceedings{rutkowski-etal-2019-estimating,
title = "Estimating senses with sets of lexically related words for {P}olish word sense disambiguation",
author = "Rutkowski, Szymon and
Rychlik, Piotr and
Mykowiecka, Agnieszka",
editor = "Vossen, Piek and
Fellbaum, Christiane",
booktitle = "Proceedings of the 10th Global Wordnet Conference",
month = jul,
year = "2019",
address = "Wroclaw, Poland",
publisher = "Global Wordnet Association",
url = "https://aclanthology.org/2019.gwc-1.15",
pages = "118--124",
abstract = "We propose a new algorithm for word sense disambiguation, exploiting data from a WordNet with many types of lexical relations, such as plWordNet for Polish. In this method, sense probabilities in context are approximated with a language model. To estimate the likelihood of a sense appearing amidst the word sequence, the token being disambiguated is substituted with words related lexically to the given sense or words appearing in its WordNet gloss. We test this approach on a set of sense-annotated Polish sentences with a number of neural language models. Our best setup achieves the accuracy score of 55.12{\%} (72.02{\%} when first senses are excluded), up from 51.77{\%} of an existing PageRank-based method. While not exceeding the first (often meaning most frequent) sense baseline in the standard case, this encourages further research on combining WordNet data with neural models.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="rutkowski-etal-2019-estimating">
<titleInfo>
<title>Estimating senses with sets of lexically related words for Polish word sense disambiguation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Szymon</namePart>
<namePart type="family">Rutkowski</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Piotr</namePart>
<namePart type="family">Rychlik</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Agnieszka</namePart>
<namePart type="family">Mykowiecka</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2019-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 10th Global Wordnet Conference</title>
</titleInfo>
<name type="personal">
<namePart type="given">Piek</namePart>
<namePart type="family">Vossen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Christiane</namePart>
<namePart type="family">Fellbaum</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Global Wordnet Association</publisher>
<place>
<placeTerm type="text">Wroclaw, Poland</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>We propose a new algorithm for word sense disambiguation, exploiting data from a WordNet with many types of lexical relations, such as plWordNet for Polish. In this method, sense probabilities in context are approximated with a language model. To estimate the likelihood of a sense appearing amidst the word sequence, the token being disambiguated is substituted with words related lexically to the given sense or words appearing in its WordNet gloss. We test this approach on a set of sense-annotated Polish sentences with a number of neural language models. Our best setup achieves the accuracy score of 55.12% (72.02% when first senses are excluded), up from 51.77% of an existing PageRank-based method. While not exceeding the first (often meaning most frequent) sense baseline in the standard case, this encourages further research on combining WordNet data with neural models.</abstract>
<identifier type="citekey">rutkowski-etal-2019-estimating</identifier>
<location>
<url>https://aclanthology.org/2019.gwc-1.15</url>
</location>
<part>
<date>2019-07</date>
<extent unit="page">
<start>118</start>
<end>124</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Estimating senses with sets of lexically related words for Polish word sense disambiguation
%A Rutkowski, Szymon
%A Rychlik, Piotr
%A Mykowiecka, Agnieszka
%Y Vossen, Piek
%Y Fellbaum, Christiane
%S Proceedings of the 10th Global Wordnet Conference
%D 2019
%8 July
%I Global Wordnet Association
%C Wroclaw, Poland
%F rutkowski-etal-2019-estimating
%X We propose a new algorithm for word sense disambiguation, exploiting data from a WordNet with many types of lexical relations, such as plWordNet for Polish. In this method, sense probabilities in context are approximated with a language model. To estimate the likelihood of a sense appearing amidst the word sequence, the token being disambiguated is substituted with words related lexically to the given sense or words appearing in its WordNet gloss. We test this approach on a set of sense-annotated Polish sentences with a number of neural language models. Our best setup achieves the accuracy score of 55.12% (72.02% when first senses are excluded), up from 51.77% of an existing PageRank-based method. While not exceeding the first (often meaning most frequent) sense baseline in the standard case, this encourages further research on combining WordNet data with neural models.
%U https://aclanthology.org/2019.gwc-1.15
%P 118-124
Markdown (Informal)
[Estimating senses with sets of lexically related words for Polish word sense disambiguation](https://aclanthology.org/2019.gwc-1.15) (Rutkowski et al., GWC 2019)
ACL