@inproceedings{hovy-etal-2014-pos,
title = "When {POS} data sets don{'}t add up: Combatting sample bias",
author = "Hovy, Dirk and
Plank, Barbara and
S{\o}gaard, Anders",
editor = "Calzolari, Nicoletta and
Choukri, Khalid and
Declerck, Thierry and
Loftsson, Hrafn and
Maegaard, Bente and
Mariani, Joseph and
Moreno, Asuncion and
Odijk, Jan and
Piperidis, Stelios",
booktitle = "Proceedings of the Ninth International Conference on Language Resources and Evaluation ({LREC}'14)",
month = may,
year = "2014",
address = "Reykjavik, Iceland",
publisher = "European Language Resources Association (ELRA)",
url = "http://www.lrec-conf.org/proceedings/lrec2014/pdf/476_Paper.pdf",
pages = "4472--4475",
abstract = "Several works in Natural Language Processing have recently looked into part-of-speech annotation of Twitter data and typically used their own data sets. Since conventions on Twitter change rapidly, models often show sample bias. Training on a combination of the existing data sets should help overcome this bias and produce more robust models than any trained on the individual corpora. Unfortunately, combining the existing corpora proves difficult: many of the corpora use proprietary tag sets that have little or no overlap. Even when mapped to a common tag set, the different corpora systematically differ in their treatment of various tags and tokens. This includes both pre-processing decisions, as well as default labels for frequent tokens, thus exhibiting data bias and label bias, respectively. Only if we address these biases can we combine the existing data sets to also overcome sample bias. We present a systematic study of several Twitter POS data sets, the problems of label and data bias, discuss their effects on model performance, and show how to overcome them to learn models that perform well on various test sets, achieving relative error reduction of up to 21{\%}.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="hovy-etal-2014-pos">
<titleInfo>
<title>When POS data sets don’t add up: Combatting sample bias</title>
</titleInfo>
<name type="personal">
<namePart type="given">Dirk</namePart>
<namePart type="family">Hovy</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Barbara</namePart>
<namePart type="family">Plank</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Anders</namePart>
<namePart type="family">Søgaard</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2014-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC’14)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Nicoletta</namePart>
<namePart type="family">Calzolari</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Khalid</namePart>
<namePart type="family">Choukri</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Thierry</namePart>
<namePart type="family">Declerck</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hrafn</namePart>
<namePart type="family">Loftsson</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bente</namePart>
<namePart type="family">Maegaard</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Joseph</namePart>
<namePart type="family">Mariani</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Asuncion</namePart>
<namePart type="family">Moreno</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jan</namePart>
<namePart type="family">Odijk</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Stelios</namePart>
<namePart type="family">Piperidis</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>European Language Resources Association (ELRA)</publisher>
<place>
<placeTerm type="text">Reykjavik, Iceland</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Several works in Natural Language Processing have recently looked into part-of-speech annotation of Twitter data and typically used their own data sets. Since conventions on Twitter change rapidly, models often show sample bias. Training on a combination of the existing data sets should help overcome this bias and produce more robust models than any trained on the individual corpora. Unfortunately, combining the existing corpora proves difficult: many of the corpora use proprietary tag sets that have little or no overlap. Even when mapped to a common tag set, the different corpora systematically differ in their treatment of various tags and tokens. This includes both pre-processing decisions, as well as default labels for frequent tokens, thus exhibiting data bias and label bias, respectively. Only if we address these biases can we combine the existing data sets to also overcome sample bias. We present a systematic study of several Twitter POS data sets, the problems of label and data bias, discuss their effects on model performance, and show how to overcome them to learn models that perform well on various test sets, achieving relative error reduction of up to 21%.</abstract>
<identifier type="citekey">hovy-etal-2014-pos</identifier>
<location>
<url>http://www.lrec-conf.org/proceedings/lrec2014/pdf/476_Paper.pdf</url>
</location>
<part>
<date>2014-05</date>
<extent unit="page">
<start>4472</start>
<end>4475</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T When POS data sets don’t add up: Combatting sample bias
%A Hovy, Dirk
%A Plank, Barbara
%A Søgaard, Anders
%Y Calzolari, Nicoletta
%Y Choukri, Khalid
%Y Declerck, Thierry
%Y Loftsson, Hrafn
%Y Maegaard, Bente
%Y Mariani, Joseph
%Y Moreno, Asuncion
%Y Odijk, Jan
%Y Piperidis, Stelios
%S Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC’14)
%D 2014
%8 May
%I European Language Resources Association (ELRA)
%C Reykjavik, Iceland
%F hovy-etal-2014-pos
%X Several works in Natural Language Processing have recently looked into part-of-speech annotation of Twitter data and typically used their own data sets. Since conventions on Twitter change rapidly, models often show sample bias. Training on a combination of the existing data sets should help overcome this bias and produce more robust models than any trained on the individual corpora. Unfortunately, combining the existing corpora proves difficult: many of the corpora use proprietary tag sets that have little or no overlap. Even when mapped to a common tag set, the different corpora systematically differ in their treatment of various tags and tokens. This includes both pre-processing decisions, as well as default labels for frequent tokens, thus exhibiting data bias and label bias, respectively. Only if we address these biases can we combine the existing data sets to also overcome sample bias. We present a systematic study of several Twitter POS data sets, the problems of label and data bias, discuss their effects on model performance, and show how to overcome them to learn models that perform well on various test sets, achieving relative error reduction of up to 21%.
%U http://www.lrec-conf.org/proceedings/lrec2014/pdf/476_Paper.pdf
%P 4472-4475
Markdown (Informal)
[When POS data sets don’t add up: Combatting sample bias](http://www.lrec-conf.org/proceedings/lrec2014/pdf/476_Paper.pdf) (Hovy et al., LREC 2014)
ACL
- Dirk Hovy, Barbara Plank, and Anders Søgaard. 2014. When POS data sets don’t add up: Combatting sample bias. In Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC'14), pages 4472–4475, Reykjavik, Iceland. European Language Resources Association (ELRA).