EMOTAG: AN APPROACH TO AUTOMATED MARKUP OF EMOTIONS IN TEXTS
Identifieur interne : 000163 ( Main/Exploration ); précédent : 000162; suivant : 000164EMOTAG: AN APPROACH TO AUTOMATED MARKUP OF EMOTIONS IN TEXTS
Auteurs : Virginia Francisco [Espagne] ; Pablo Gervás [Espagne]Source :
- Computational Intelligence [ 0824-7935 ] ; 2013-11.
English descriptors
- Teeft :
- Activation dimension, Adjective, Affective, Affective dictionaries, Affective dictionary, Affective label, Algorithm, Aman, Annotation, Annual meeting, Automatic markup, Available databases, Average value, Baseline, Basic emotions, Basic version, Bestgen, Blog, Blog posts, Boucouvalas, Brief outline, Category, Collocation, Collocation factors, Computational, Computational intelligence, Computational intelligence figure, Computational linguistics, Connotation, Construction process, Corpus, Database, Different aspects, Different emotions, Different evaluators, Different levels, Different words, Dimension, Disgust, Emotag, Emotag obtains, Emotion, Emotional, Emotional categories, Emotional category, Emotional concept, Emotional concepts, Emotional connotation, Emotional connotations, Emotional content, Emotional dimension, Emotional dimensions, Emotional information, Emotional markup, Emotional ontology, Emotional states, Emotional value, Emotional values, Emotional words, Essential aspects, Etzioni, Evaluation dimension, Evaluation measurement, Evaluator, Fairy tales, Future work, General dictionary, General inquirer, Generic, Generic emotions, Headline, Human evaluators, Hypernym, Information retrieval, Language resources, Large number, Lasswell value dictionary, Last sentence, Least half, Lexical, Lexical labels, Main sentence, Markup, Markup process, Metrics, Mihalcea, Naive bayes, Narrative texts, Natural language, Natural language processing, Negation, Neutral value, Noun, Numerical values, Ontology, Opposite emotion, Opposite emotions, Ortony, Other emotions, Pang, Particular occurrence, Pearson measure, Popescu, Power dimension, Power dimensions, Previous work, Reference value, Research efforts, Resource, Sadness, Secondary emotions, Semantic orientation, Semeval, Sentence, Sentiment analysis, Single element, Social psychology, Sproat, Strapparava, Subjectivity, Subordinate clause, Subordinate clauses, Subordinate sentences, Success rate, Sugimoto, Synset, Syntactic structure, System marks, Szpakowicz, Tagger, Tertiary emotions, Test collection, Turney, Valence, Wiebe, Wordnet, Yoneyama.
Abstract
This paper presents an approach to the automated markup of texts with emotional labels. The approach considers two possible representations of emotions in parallel: emotional categories (emotional tags used to refer to emotions) and emotional dimensions (measures that try to model the essential aspects of emotions numerically). For each representation, a corpus of example texts previously annotated by human evaluators is mined for an initial assignment of emotional features to words. This results in a list of emotional words (LEW) which becomes a useful resource for later automated markup. The algorithm proposed for the automated markup of text closely mirrors the steps taken during feature extraction, employing a combination of the LEW resource and the ANEW word list for the actual assignment of emotional features, and WordNet for knowledge‐based expansion of words not occurring in either and an ontology of emotional categories. The algorithm for automated markup is tested and the results are discussed with respect to three main issues: the relative adequacy of each of the representations used, correctness and coverage of the proposed algorithm, and additional techniques and solutions that may be employed to improve the results. The average percentage of success obtained by our approach when it marks up with emotional dimensions is around 80% and when it marks up with emotional categories is around 50%. The main contribution of the approach presented in this paper is that it allows dimensions and categories at different levels of abstraction to operate simultaneously during markup.
Url:
DOI: 10.1111/j.1467-8640.2012.00438.x
Affiliations:
Links toward previous steps (curation, corpus...)
- to stream Istex, to step Corpus: 000D24
- to stream Istex, to step Curation: 000C59
- to stream Istex, to step Checkpoint: 000048
- to stream Main, to step Merge: 000163
- to stream Main, to step Curation: 000163
Le document en format XML
<record><TEI wicri:istexFullTextTei="biblStruct"><teiHeader><fileDesc><titleStmt><title xml:lang="en">EMOTAG: AN APPROACH TO AUTOMATED MARKUP OF EMOTIONS IN TEXTS</title><author><name sortKey="Francisco, Virginia" sort="Francisco, Virginia" uniqKey="Francisco V" first="Virginia" last="Francisco">Virginia Francisco</name></author><author><name sortKey="Gervas, Pablo" sort="Gervas, Pablo" uniqKey="Gervas P" first="Pablo" last="Gervás">Pablo Gervás</name></author></titleStmt><publicationStmt><idno type="wicri:source">ISTEX</idno><idno type="RBID">ISTEX:7EB0098D796B2DAFEA9D213EB0FB0AF7CD161504</idno><date when="2013" year="2013">2013</date><idno type="doi">10.1111/j.1467-8640.2012.00438.x</idno><idno type="url">https://api.istex.fr/document/7EB0098D796B2DAFEA9D213EB0FB0AF7CD161504/fulltext/pdf</idno><idno type="wicri:Area/Istex/Corpus">000D24</idno><idno type="wicri:explorRef" wicri:stream="Istex" wicri:step="Corpus" wicri:corpus="ISTEX">000D24</idno><idno type="wicri:Area/Istex/Curation">000C59</idno><idno type="wicri:Area/Istex/Checkpoint">000048</idno><idno type="wicri:explorRef" wicri:stream="Istex" wicri:step="Checkpoint">000048</idno><idno type="wicri:doubleKey">0824-7935:2013:Francisco V:emotag:an:approach</idno><idno type="wicri:Area/Main/Merge">000163</idno><idno type="wicri:Area/Main/Curation">000163</idno><idno type="wicri:Area/Main/Exploration">000163</idno></publicationStmt><sourceDesc><biblStruct><analytic><title level="a" type="main" xml:lang="en">EMOTAG: AN APPROACH TO AUTOMATED MARKUP OF EMOTIONS IN TEXTS</title><author><name sortKey="Francisco, Virginia" sort="Francisco, Virginia" uniqKey="Francisco V" first="Virginia" last="Francisco">Virginia Francisco</name><affiliation wicri:level="4"><country xml:lang="fr">Espagne</country><wicri:regionArea>Departamento de Ingeniería del Software e Inteligencia Artificial, Universidad Complutense de Madrid, Madrid</wicri:regionArea><placeName><settlement type="city">Madrid</settlement><region nuts="2" type="region">Communauté de Madrid</region></placeName><orgName type="university">Université complutense de Madrid</orgName></affiliation><affiliation wicri:level="1"><country wicri:rule="url">Espagne</country></affiliation></author><author><name sortKey="Gervas, Pablo" sort="Gervas, Pablo" uniqKey="Gervas P" first="Pablo" last="Gervás">Pablo Gervás</name><affiliation wicri:level="4"><country xml:lang="fr">Espagne</country><wicri:regionArea>Instituto de Tecnología del Conocimiento, Universidad Complutense de Madrid, Madrid</wicri:regionArea><placeName><settlement type="city">Madrid</settlement><region nuts="2" type="region">Communauté de Madrid</region></placeName><orgName type="university">Université complutense de Madrid</orgName></affiliation></author></analytic><monogr></monogr><series><title level="j">Computational Intelligence</title><title level="j" type="abbrev">Computational Intelligence</title><idno type="ISSN">0824-7935</idno><idno type="eISSN">1467-8640</idno><imprint><publisher>Blackwell Publishing Ltd</publisher><date type="published" when="2013-11">2013-11</date><biblScope unit="volume">29</biblScope><biblScope unit="issue">4</biblScope><biblScope unit="page" from="680">680</biblScope><biblScope unit="page" to="721">721</biblScope></imprint><idno type="ISSN">0824-7935</idno></series><idno type="istex">7EB0098D796B2DAFEA9D213EB0FB0AF7CD161504</idno><idno type="DOI">10.1111/j.1467-8640.2012.00438.x</idno><idno type="ArticleID">COIN438</idno></biblStruct></sourceDesc><seriesStmt><idno type="ISSN">0824-7935</idno></seriesStmt></fileDesc><profileDesc><textClass><keywords scheme="Teeft" xml:lang="en"><term>Activation dimension</term><term>Adjective</term><term>Affective</term><term>Affective dictionaries</term><term>Affective dictionary</term><term>Affective label</term><term>Algorithm</term><term>Aman</term><term>Annotation</term><term>Annual meeting</term><term>Automatic markup</term><term>Available databases</term><term>Average value</term><term>Baseline</term><term>Basic emotions</term><term>Basic version</term><term>Bestgen</term><term>Blog</term><term>Blog posts</term><term>Boucouvalas</term><term>Brief outline</term><term>Category</term><term>Collocation</term><term>Collocation factors</term><term>Computational</term><term>Computational intelligence</term><term>Computational intelligence figure</term><term>Computational linguistics</term><term>Connotation</term><term>Construction process</term><term>Corpus</term><term>Database</term><term>Different aspects</term><term>Different emotions</term><term>Different evaluators</term><term>Different levels</term><term>Different words</term><term>Dimension</term><term>Disgust</term><term>Emotag</term><term>Emotag obtains</term><term>Emotion</term><term>Emotional</term><term>Emotional categories</term><term>Emotional category</term><term>Emotional concept</term><term>Emotional concepts</term><term>Emotional connotation</term><term>Emotional connotations</term><term>Emotional content</term><term>Emotional dimension</term><term>Emotional dimensions</term><term>Emotional information</term><term>Emotional markup</term><term>Emotional ontology</term><term>Emotional states</term><term>Emotional value</term><term>Emotional values</term><term>Emotional words</term><term>Essential aspects</term><term>Etzioni</term><term>Evaluation dimension</term><term>Evaluation measurement</term><term>Evaluator</term><term>Fairy tales</term><term>Future work</term><term>General dictionary</term><term>General inquirer</term><term>Generic</term><term>Generic emotions</term><term>Headline</term><term>Human evaluators</term><term>Hypernym</term><term>Information retrieval</term><term>Language resources</term><term>Large number</term><term>Lasswell value dictionary</term><term>Last sentence</term><term>Least half</term><term>Lexical</term><term>Lexical labels</term><term>Main sentence</term><term>Markup</term><term>Markup process</term><term>Metrics</term><term>Mihalcea</term><term>Naive bayes</term><term>Narrative texts</term><term>Natural language</term><term>Natural language processing</term><term>Negation</term><term>Neutral value</term><term>Noun</term><term>Numerical values</term><term>Ontology</term><term>Opposite emotion</term><term>Opposite emotions</term><term>Ortony</term><term>Other emotions</term><term>Pang</term><term>Particular occurrence</term><term>Pearson measure</term><term>Popescu</term><term>Power dimension</term><term>Power dimensions</term><term>Previous work</term><term>Reference value</term><term>Research efforts</term><term>Resource</term><term>Sadness</term><term>Secondary emotions</term><term>Semantic orientation</term><term>Semeval</term><term>Sentence</term><term>Sentiment analysis</term><term>Single element</term><term>Social psychology</term><term>Sproat</term><term>Strapparava</term><term>Subjectivity</term><term>Subordinate clause</term><term>Subordinate clauses</term><term>Subordinate sentences</term><term>Success rate</term><term>Sugimoto</term><term>Synset</term><term>Syntactic structure</term><term>System marks</term><term>Szpakowicz</term><term>Tagger</term><term>Tertiary emotions</term><term>Test collection</term><term>Turney</term><term>Valence</term><term>Wiebe</term><term>Wordnet</term><term>Yoneyama</term></keywords></textClass><langUsage><language ident="en">en</language></langUsage></profileDesc></teiHeader><front><div type="abstract">This paper presents an approach to the automated markup of texts with emotional labels. The approach considers two possible representations of emotions in parallel: emotional categories (emotional tags used to refer to emotions) and emotional dimensions (measures that try to model the essential aspects of emotions numerically). For each representation, a corpus of example texts previously annotated by human evaluators is mined for an initial assignment of emotional features to words. This results in a list of emotional words (LEW) which becomes a useful resource for later automated markup. The algorithm proposed for the automated markup of text closely mirrors the steps taken during feature extraction, employing a combination of the LEW resource and the ANEW word list for the actual assignment of emotional features, and WordNet for knowledge‐based expansion of words not occurring in either and an ontology of emotional categories. The algorithm for automated markup is tested and the results are discussed with respect to three main issues: the relative adequacy of each of the representations used, correctness and coverage of the proposed algorithm, and additional techniques and solutions that may be employed to improve the results. The average percentage of success obtained by our approach when it marks up with emotional dimensions is around 80% and when it marks up with emotional categories is around 50%. The main contribution of the approach presented in this paper is that it allows dimensions and categories at different levels of abstraction to operate simultaneously during markup.</div></front></TEI><affiliations><list><country><li>Espagne</li></country><region><li>Communauté de Madrid</li></region><settlement><li>Madrid</li></settlement><orgName><li>Université complutense de Madrid</li></orgName></list><tree><country name="Espagne"><region name="Communauté de Madrid"><name sortKey="Francisco, Virginia" sort="Francisco, Virginia" uniqKey="Francisco V" first="Virginia" last="Francisco">Virginia Francisco</name></region><name sortKey="Francisco, Virginia" sort="Francisco, Virginia" uniqKey="Francisco V" first="Virginia" last="Francisco">Virginia Francisco</name><name sortKey="Gervas, Pablo" sort="Gervas, Pablo" uniqKey="Gervas P" first="Pablo" last="Gervás">Pablo Gervás</name></country></tree></affiliations></record>Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Wicri/Sarre/explor/MusicSarreV3/Data/Main/Exploration
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 000163 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 000163 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Wicri/Sarre
|area= MusicSarreV3
|flux= Main
|étape= Exploration
|type= RBID
|clé= ISTEX:7EB0098D796B2DAFEA9D213EB0FB0AF7CD161504
|texte= EMOTAG: AN APPROACH TO AUTOMATED MARKUP OF EMOTIONS IN TEXTS
}}
| This area was generated with Dilib version V0.6.33. |  |