@inproceedings{schneidermann-etal-2020-towards,
title = "Towards a Gold Standard for Evaluating {D}anish Word Embeddings",
author = "Schneidermann, Nina and
Hvingelby, Rasmus and
Pedersen, Bolette",
booktitle = "Proceedings of the 12th Language Resources and Evaluation Conference",
month = may,
year = "2020",
address = "Marseille, France",
publisher = "European Language Resources Association",
url = "https://aclanthology.org/2020.lrec-1.585",
pages = "4754--4763",
abstract = "This paper presents the process of compiling a model-agnostic similarity goal standard for evaluating Danish word embeddings based on human judgments made by 42 native speakers of Danish. Word embeddings resemble semantic similarity solely by distribution (meaning that word vectors do not reflect relatedness as differing from similarity), and we argue that this generalization poses a problem in most intrinsic evaluation scenarios. In order to be able to evaluate on both dimensions, our human-generated dataset is therefore designed to reflect the distinction between relatedness and similarity. The goal standard is applied for evaluating the {``}goodness{''} of six existing word embedding models for Danish, and it is discussed how a relatively low correlation can be explained by the fact that semantic similarity is substantially more challenging to model than relatedness, and that there seems to be a need for future human judgments to measure similarity in full context and along more than a single spectrum.",
language = "English",
ISBN = "979-10-95546-34-4",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="schneidermann-etal-2020-towards">
<titleInfo>
<title>Towards a Gold Standard for Evaluating Danish Word Embeddings</title>
</titleInfo>
<name type="personal">
<namePart type="given">Nina</namePart>
<namePart type="family">Schneidermann</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rasmus</namePart>
<namePart type="family">Hvingelby</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bolette</namePart>
<namePart type="family">Pedersen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2020-may</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<language>
<languageTerm type="text">English</languageTerm>
<languageTerm type="code" authority="iso639-2b">eng</languageTerm>
</language>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 12th Language Resources and Evaluation Conference</title>
</titleInfo>
<originInfo>
<publisher>European Language Resources Association</publisher>
<place>
<placeTerm type="text">Marseille, France</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-10-95546-34-4</identifier>
</relatedItem>
<abstract>This paper presents the process of compiling a model-agnostic similarity goal standard for evaluating Danish word embeddings based on human judgments made by 42 native speakers of Danish. Word embeddings resemble semantic similarity solely by distribution (meaning that word vectors do not reflect relatedness as differing from similarity), and we argue that this generalization poses a problem in most intrinsic evaluation scenarios. In order to be able to evaluate on both dimensions, our human-generated dataset is therefore designed to reflect the distinction between relatedness and similarity. The goal standard is applied for evaluating the “goodness” of six existing word embedding models for Danish, and it is discussed how a relatively low correlation can be explained by the fact that semantic similarity is substantially more challenging to model than relatedness, and that there seems to be a need for future human judgments to measure similarity in full context and along more than a single spectrum.</abstract>
<identifier type="citekey">schneidermann-etal-2020-towards</identifier>
<location>
<url>https://aclanthology.org/2020.lrec-1.585</url>
</location>
<part>
<date>2020-may</date>
<extent unit="page">
<start>4754</start>
<end>4763</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Towards a Gold Standard for Evaluating Danish Word Embeddings
%A Schneidermann, Nina
%A Hvingelby, Rasmus
%A Pedersen, Bolette
%S Proceedings of the 12th Language Resources and Evaluation Conference
%D 2020
%8 may
%I European Language Resources Association
%C Marseille, France
%@ 979-10-95546-34-4
%G English
%F schneidermann-etal-2020-towards
%X This paper presents the process of compiling a model-agnostic similarity goal standard for evaluating Danish word embeddings based on human judgments made by 42 native speakers of Danish. Word embeddings resemble semantic similarity solely by distribution (meaning that word vectors do not reflect relatedness as differing from similarity), and we argue that this generalization poses a problem in most intrinsic evaluation scenarios. In order to be able to evaluate on both dimensions, our human-generated dataset is therefore designed to reflect the distinction between relatedness and similarity. The goal standard is applied for evaluating the “goodness” of six existing word embedding models for Danish, and it is discussed how a relatively low correlation can be explained by the fact that semantic similarity is substantially more challenging to model than relatedness, and that there seems to be a need for future human judgments to measure similarity in full context and along more than a single spectrum.
%U https://aclanthology.org/2020.lrec-1.585
%P 4754-4763
Markdown (Informal)
[Towards a Gold Standard for Evaluating Danish Word Embeddings](https://aclanthology.org/2020.lrec-1.585) (Schneidermann et al., LREC 2020)
ACL