@article{smith-etal-2017-evaluating,
title = "Evaluating Visual Representations for Topic Understanding and Their Effects on Manually Generated Topic Labels",
author = "Smith, Alison and
Lee, Tak Yeon and
Poursabzi-Sangdeh, Forough and
Boyd-Graber, Jordan and
Elmqvist, Niklas and
Findlater, Leah",
editor = "Lee, Lillian and
Johnson, Mark and
Toutanova, Kristina",
journal = "Transactions of the Association for Computational Linguistics",
volume = "5",
year = "2017",
address = "Cambridge, MA",
publisher = "MIT Press",
url = "https://aclanthology.org/Q17-1001",
doi = "10.1162/tacl_a_00042",
pages = "1--16",
abstract = "Probabilistic topic models are important tools for indexing, summarizing, and analyzing large document collections by their themes. However, promoting end-user understanding of topics remains an open research problem. We compare labels generated by users given four topic visualization techniques{---}word lists, word lists with bars, word clouds, and network graphs{---}against each other and against automatically generated labels. Our basis of comparison is participant ratings of how well labels describe documents from the topic. Our study has two phases: a labeling phase where participants label visualized topics and a validation phase where different participants select which labels best describe the topics{'} documents. Although all visualizations produce similar quality labels, simple visualizations such as word lists allow participants to quickly understand topics, while complex visualizations take longer but expose multi-word expressions that simpler visualizations obscure. Automatic labels lag behind user-created labels, but our dataset of manually labeled topics highlights linguistic patterns (e.g., hypernyms, phrases) that can be used to improve automatic topic labeling algorithms.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="smith-etal-2017-evaluating">
<titleInfo>
<title>Evaluating Visual Representations for Topic Understanding and Their Effects on Manually Generated Topic Labels</title>
</titleInfo>
<name type="personal">
<namePart type="given">Alison</namePart>
<namePart type="family">Smith</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tak</namePart>
<namePart type="given">Yeon</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Forough</namePart>
<namePart type="family">Poursabzi-Sangdeh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jordan</namePart>
<namePart type="family">Boyd-Graber</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Niklas</namePart>
<namePart type="family">Elmqvist</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Leah</namePart>
<namePart type="family">Findlater</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2017</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<genre authority="bibutilsgt">journal article</genre>
<relatedItem type="host">
<titleInfo>
<title>Transactions of the Association for Computational Linguistics</title>
</titleInfo>
<originInfo>
<issuance>continuing</issuance>
<publisher>MIT Press</publisher>
<place>
<placeTerm type="text">Cambridge, MA</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">periodical</genre>
<genre authority="bibutilsgt">academic journal</genre>
</relatedItem>
<abstract>Probabilistic topic models are important tools for indexing, summarizing, and analyzing large document collections by their themes. However, promoting end-user understanding of topics remains an open research problem. We compare labels generated by users given four topic visualization techniques—word lists, word lists with bars, word clouds, and network graphs—against each other and against automatically generated labels. Our basis of comparison is participant ratings of how well labels describe documents from the topic. Our study has two phases: a labeling phase where participants label visualized topics and a validation phase where different participants select which labels best describe the topics’ documents. Although all visualizations produce similar quality labels, simple visualizations such as word lists allow participants to quickly understand topics, while complex visualizations take longer but expose multi-word expressions that simpler visualizations obscure. Automatic labels lag behind user-created labels, but our dataset of manually labeled topics highlights linguistic patterns (e.g., hypernyms, phrases) that can be used to improve automatic topic labeling algorithms.</abstract>
<identifier type="citekey">smith-etal-2017-evaluating</identifier>
<identifier type="doi">10.1162/tacl_a_00042</identifier>
<location>
<url>https://aclanthology.org/Q17-1001</url>
</location>
<part>
<date>2017</date>
<detail type="volume"><number>5</number></detail>
<extent unit="page">
<start>1</start>
<end>16</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Journal Article
%T Evaluating Visual Representations for Topic Understanding and Their Effects on Manually Generated Topic Labels
%A Smith, Alison
%A Lee, Tak Yeon
%A Poursabzi-Sangdeh, Forough
%A Boyd-Graber, Jordan
%A Elmqvist, Niklas
%A Findlater, Leah
%J Transactions of the Association for Computational Linguistics
%D 2017
%V 5
%I MIT Press
%C Cambridge, MA
%F smith-etal-2017-evaluating
%X Probabilistic topic models are important tools for indexing, summarizing, and analyzing large document collections by their themes. However, promoting end-user understanding of topics remains an open research problem. We compare labels generated by users given four topic visualization techniques—word lists, word lists with bars, word clouds, and network graphs—against each other and against automatically generated labels. Our basis of comparison is participant ratings of how well labels describe documents from the topic. Our study has two phases: a labeling phase where participants label visualized topics and a validation phase where different participants select which labels best describe the topics’ documents. Although all visualizations produce similar quality labels, simple visualizations such as word lists allow participants to quickly understand topics, while complex visualizations take longer but expose multi-word expressions that simpler visualizations obscure. Automatic labels lag behind user-created labels, but our dataset of manually labeled topics highlights linguistic patterns (e.g., hypernyms, phrases) that can be used to improve automatic topic labeling algorithms.
%R 10.1162/tacl_a_00042
%U https://aclanthology.org/Q17-1001
%U https://doi.org/10.1162/tacl_a_00042
%P 1-16
Markdown (Informal)
[Evaluating Visual Representations for Topic Understanding and Their Effects on Manually Generated Topic Labels](https://aclanthology.org/Q17-1001) (Smith et al., TACL 2017)
ACL