@inproceedings{passonneau-etal-2008-relation,
title = "Relation between Agreement Measures on Human Labeling and Machine Learning Performance: Results from an Art History Domain",
author = "Passonneau, Rebecca and
Lippincott, Tom and
Yano, Tae and
Klavans, Judith",
editor = "Calzolari, Nicoletta and
Choukri, Khalid and
Maegaard, Bente and
Mariani, Joseph and
Odijk, Jan and
Piperidis, Stelios and
Tapias, Daniel",
booktitle = "Proceedings of the Sixth International Conference on Language Resources and Evaluation ({LREC}'08)",
month = may,
year = "2008",
address = "Marrakech, Morocco",
publisher = "European Language Resources Association (ELRA)",
url = "http://www.lrec-conf.org/proceedings/lrec2008/pdf/722_paper.pdf",
abstract = "We discuss factors that affect human agreement on a semantic labeling task in the art history domain, based on the results of four experiments where we varied the number of labels annotators could assign, the number of annotators, the type and amount of training they received, and the size of the text span being labeled. Using the labelings from one experiment involving seven annotators, we investigate the relation between interannotator agreement and machine learning performance. We construct binary classifiers and vary the training and test data by swapping the labelings from the seven annotators. First, we find performance is often quite good despite lower than recommended interannotator agreement. Second, we find that on average, learning performance for a given functional semantic category correlates with the overall agreement among the seven annotators for that category. Third, we find that learning performance on the data from a given annotator does not correlate with the quality of that annotators labeling. We offer recommendations for the use of labeled data in machine learning, and argue that learners should attempt to accommodate human variation. We also note implications for large scale corpus annotation projects that deal with similarly subjective phenomena.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="passonneau-etal-2008-relation">
<titleInfo>
<title>Relation between Agreement Measures on Human Labeling and Machine Learning Performance: Results from an Art History Domain</title>
</titleInfo>
<name type="personal">
<namePart type="given">Rebecca</namePart>
<namePart type="family">Passonneau</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tom</namePart>
<namePart type="family">Lippincott</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tae</namePart>
<namePart type="family">Yano</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Judith</namePart>
<namePart type="family">Klavans</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2008-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Sixth International Conference on Language Resources and Evaluation (LREC’08)</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">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">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>
<name type="personal">
<namePart type="given">Daniel</namePart>
<namePart type="family">Tapias</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>European Language Resources Association (ELRA)</publisher>
<place>
<placeTerm type="text">Marrakech, Morocco</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>We discuss factors that affect human agreement on a semantic labeling task in the art history domain, based on the results of four experiments where we varied the number of labels annotators could assign, the number of annotators, the type and amount of training they received, and the size of the text span being labeled. Using the labelings from one experiment involving seven annotators, we investigate the relation between interannotator agreement and machine learning performance. We construct binary classifiers and vary the training and test data by swapping the labelings from the seven annotators. First, we find performance is often quite good despite lower than recommended interannotator agreement. Second, we find that on average, learning performance for a given functional semantic category correlates with the overall agreement among the seven annotators for that category. Third, we find that learning performance on the data from a given annotator does not correlate with the quality of that annotators labeling. We offer recommendations for the use of labeled data in machine learning, and argue that learners should attempt to accommodate human variation. We also note implications for large scale corpus annotation projects that deal with similarly subjective phenomena.</abstract>
<identifier type="citekey">passonneau-etal-2008-relation</identifier>
<location>
<url>http://www.lrec-conf.org/proceedings/lrec2008/pdf/722_paper.pdf</url>
</location>
<part>
<date>2008-05</date>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Relation between Agreement Measures on Human Labeling and Machine Learning Performance: Results from an Art History Domain
%A Passonneau, Rebecca
%A Lippincott, Tom
%A Yano, Tae
%A Klavans, Judith
%Y Calzolari, Nicoletta
%Y Choukri, Khalid
%Y Maegaard, Bente
%Y Mariani, Joseph
%Y Odijk, Jan
%Y Piperidis, Stelios
%Y Tapias, Daniel
%S Proceedings of the Sixth International Conference on Language Resources and Evaluation (LREC’08)
%D 2008
%8 May
%I European Language Resources Association (ELRA)
%C Marrakech, Morocco
%F passonneau-etal-2008-relation
%X We discuss factors that affect human agreement on a semantic labeling task in the art history domain, based on the results of four experiments where we varied the number of labels annotators could assign, the number of annotators, the type and amount of training they received, and the size of the text span being labeled. Using the labelings from one experiment involving seven annotators, we investigate the relation between interannotator agreement and machine learning performance. We construct binary classifiers and vary the training and test data by swapping the labelings from the seven annotators. First, we find performance is often quite good despite lower than recommended interannotator agreement. Second, we find that on average, learning performance for a given functional semantic category correlates with the overall agreement among the seven annotators for that category. Third, we find that learning performance on the data from a given annotator does not correlate with the quality of that annotators labeling. We offer recommendations for the use of labeled data in machine learning, and argue that learners should attempt to accommodate human variation. We also note implications for large scale corpus annotation projects that deal with similarly subjective phenomena.
%U http://www.lrec-conf.org/proceedings/lrec2008/pdf/722_paper.pdf
Markdown (Informal)
[Relation between Agreement Measures on Human Labeling and Machine Learning Performance: Results from an Art History Domain](http://www.lrec-conf.org/proceedings/lrec2008/pdf/722_paper.pdf) (Passonneau et al., LREC 2008)
ACL