@inproceedings{losing-etal-2021-extraction,
title = "Extraction of Common-Sense Relations from Procedural Task Instructions using {BERT}",
author = {Losing, Viktor and
Fischer, Lydia and
Deigm{\"o}ller, J{\"o}rg},
editor = "Vossen, Piek and
Fellbaum, Christiane",
booktitle = "Proceedings of the 11th Global Wordnet Conference",
month = jan,
year = "2021",
address = "University of South Africa (UNISA)",
publisher = "Global Wordnet Association",
url = "https://aclanthology.org/2021.gwc-1.10",
pages = "81--90",
abstract = "Manipulation-relevant common-sense knowledge is crucial to support action-planning for complex tasks. In particular, instrumentality information of what can be done with certain tools can be used to limit the search space which is growing exponentially with the number of viable options. Typical sources for such knowledge, structured common-sense knowledge bases such as ConceptNet or WebChild, provide a limited amount of information which also varies drastically across different domains. Considering the recent success of pre-trained language models such as BERT, we investigate whether common-sense information can directly be extracted from semi-structured text with an acceptable annotation effort. Concretely, we compare the common-sense relations obtained from ConceptNet versus those extracted with BERT from large recipe databases. In this context, we propose a scoring function, based on the WordNet taxonomy to match specific terms to more general ones, enabling a rich evaluation against a set of ground-truth relations.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="losing-etal-2021-extraction">
<titleInfo>
<title>Extraction of Common-Sense Relations from Procedural Task Instructions using BERT</title>
</titleInfo>
<name type="personal">
<namePart type="given">Viktor</namePart>
<namePart type="family">Losing</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lydia</namePart>
<namePart type="family">Fischer</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jörg</namePart>
<namePart type="family">Deigmöller</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2021-01</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 11th 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">University of South Africa (UNISA)</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Manipulation-relevant common-sense knowledge is crucial to support action-planning for complex tasks. In particular, instrumentality information of what can be done with certain tools can be used to limit the search space which is growing exponentially with the number of viable options. Typical sources for such knowledge, structured common-sense knowledge bases such as ConceptNet or WebChild, provide a limited amount of information which also varies drastically across different domains. Considering the recent success of pre-trained language models such as BERT, we investigate whether common-sense information can directly be extracted from semi-structured text with an acceptable annotation effort. Concretely, we compare the common-sense relations obtained from ConceptNet versus those extracted with BERT from large recipe databases. In this context, we propose a scoring function, based on the WordNet taxonomy to match specific terms to more general ones, enabling a rich evaluation against a set of ground-truth relations.</abstract>
<identifier type="citekey">losing-etal-2021-extraction</identifier>
<location>
<url>https://aclanthology.org/2021.gwc-1.10</url>
</location>
<part>
<date>2021-01</date>
<extent unit="page">
<start>81</start>
<end>90</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Extraction of Common-Sense Relations from Procedural Task Instructions using BERT
%A Losing, Viktor
%A Fischer, Lydia
%A Deigmöller, Jörg
%Y Vossen, Piek
%Y Fellbaum, Christiane
%S Proceedings of the 11th Global Wordnet Conference
%D 2021
%8 January
%I Global Wordnet Association
%C University of South Africa (UNISA)
%F losing-etal-2021-extraction
%X Manipulation-relevant common-sense knowledge is crucial to support action-planning for complex tasks. In particular, instrumentality information of what can be done with certain tools can be used to limit the search space which is growing exponentially with the number of viable options. Typical sources for such knowledge, structured common-sense knowledge bases such as ConceptNet or WebChild, provide a limited amount of information which also varies drastically across different domains. Considering the recent success of pre-trained language models such as BERT, we investigate whether common-sense information can directly be extracted from semi-structured text with an acceptable annotation effort. Concretely, we compare the common-sense relations obtained from ConceptNet versus those extracted with BERT from large recipe databases. In this context, we propose a scoring function, based on the WordNet taxonomy to match specific terms to more general ones, enabling a rich evaluation against a set of ground-truth relations.
%U https://aclanthology.org/2021.gwc-1.10
%P 81-90
Markdown (Informal)
[Extraction of Common-Sense Relations from Procedural Task Instructions using BERT](https://aclanthology.org/2021.gwc-1.10) (Losing et al., GWC 2021)
ACL