Contextual information usage for the enhancement of basic emotion classification in a weakly labelled social network dataset in Spanish

Basic emotion classification is one of the main tasks of Sentiment Analysis usuallyperformed by using several machine learning techniques. One of the main issues inSentiment Analysis is the availability of tagged resources to properly train super-vised classification algorithms. This is of particula...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Tessore, Juan Pablo, Esnaola, Leonardo Martín, Ramón, Hugo Dionisio, Lanzarini, Laura, Baldassarri, Sandra
Otros Autores: 0000-0002-2111-0976
Formato: Artículo acceptedVersion
Lenguaje:Inglés
Publicado: Springer Science+Business Media, LLC, part of Springer Nature 2023
Materias:
Distant supervision
Basic emotion classification
Contextual information
Social media
Acceso en línea:http://repositorio.unnoba.edu.ar/xmlui/handle/23601/645
Aporte de:
Re DI Repositorio Digital UNNOBA de Universidad Nacional del Noroeste de la Provincia de Buenos Aires
id I103-R405-23601-645
record_format dspace
institution Universidad Nacional del Noroeste de la Provincia de Buenos Aires
institution_str I-103
repository_str R-405
collection Re DI Repositorio Digital UNNOBA
language Inglés
topic Distant supervision
Basic emotion classification
Contextual information
Social media
spellingShingle Distant supervision
Basic emotion classification
Contextual information
Social media
Tessore, Juan Pablo
Esnaola, Leonardo Martín
Ramón, Hugo Dionisio
Lanzarini, Laura
Baldassarri, Sandra
Contextual information usage for the enhancement of basic emotion classification in a weakly labelled social network dataset in Spanish
topic_facet Distant supervision
Basic emotion classification
Contextual information
Social media
description Basic emotion classification is one of the main tasks of Sentiment Analysis usuallyperformed by using several machine learning techniques. One of the main issues inSentiment Analysis is the availability of tagged resources to properly train super-vised classification algorithms. This is of particular concern in languages other thanEnglish, such as Spanish, where scarcity of these resources is the norm. In addition,most basic emotion datasets available in Spanish are rather small, containing a few hundred (or thousand) samples. Usually, the samples only contain a short text(frequently a comment) and a tag (the basic emotion), omitting crucial contextualinformation that may help to improve the classification task results. In this paper, theimpact of using contextual information is measured on a recently published Spanishbasic emotion dataset and the baseline architecture proposed in the Semantic Eval-uation 2019 competition. This particular dataset has two main advantages for thispaper. First, it was compiled using Distant Supervision and as a result it containsseveral hundred thousand samples. Secondly, the authors included valuable contex-tual information for each comment. The results show that contextual information,such as news headlines or summaries, helps improve the classification accuracy overa dataset of distantly supervised basic emotion labelled comments.
author2 0000-0002-2111-0976
author_facet 0000-0002-2111-0976
Tessore, Juan Pablo
Esnaola, Leonardo Martín
Ramón, Hugo Dionisio
Lanzarini, Laura
Baldassarri, Sandra
format Artículo
Artículo
acceptedVersion
Artículo
Artículo
acceptedVersion
Artículo
Artículo
acceptedVersion
author Tessore, Juan Pablo
Esnaola, Leonardo Martín
Ramón, Hugo Dionisio
Lanzarini, Laura
Baldassarri, Sandra
author_sort Tessore, Juan Pablo
title Contextual information usage for the enhancement of basic emotion classification in a weakly labelled social network dataset in Spanish
title_short Contextual information usage for the enhancement of basic emotion classification in a weakly labelled social network dataset in Spanish
title_full Contextual information usage for the enhancement of basic emotion classification in a weakly labelled social network dataset in Spanish
title_fullStr Contextual information usage for the enhancement of basic emotion classification in a weakly labelled social network dataset in Spanish
title_full_unstemmed Contextual information usage for the enhancement of basic emotion classification in a weakly labelled social network dataset in Spanish
title_sort contextual information usage for the enhancement of basic emotion classification in a weakly labelled social network dataset in spanish
publisher Springer Science+Business Media, LLC, part of Springer Nature
publishDate 2023
url http://repositorio.unnoba.edu.ar/xmlui/handle/23601/645
work_keys_str_mv AT tessorejuanpablo contextualinformationusagefortheenhancementofbasicemotionclassificationinaweaklylabelledsocialnetworkdatasetinspanish
AT esnaolaleonardomartin contextualinformationusagefortheenhancementofbasicemotionclassificationinaweaklylabelledsocialnetworkdatasetinspanish
AT ramonhugodionisio contextualinformationusagefortheenhancementofbasicemotionclassificationinaweaklylabelledsocialnetworkdatasetinspanish
AT lanzarinilaura contextualinformationusagefortheenhancementofbasicemotionclassificationinaweaklylabelledsocialnetworkdatasetinspanish
AT baldassarrisandra contextualinformationusagefortheenhancementofbasicemotionclassificationinaweaklylabelledsocialnetworkdatasetinspanish
_version_ 1850060746201759744
spelling I103-R405-23601-6452023-12-07T20:38:14Z Contextual information usage for the enhancement of basic emotion classification in a weakly labelled social network dataset in Spanish Tessore, Juan Pablo Esnaola, Leonardo Martín Ramón, Hugo Dionisio Lanzarini, Laura Baldassarri, Sandra 0000-0002-2111-0976 0000-0001-6298-9019 0000-0003-1577-3092 0000-0001-7027-7564 0000-0002-9315-6391 Distant supervision Basic emotion classification Contextual information Social media Basic emotion classification is one of the main tasks of Sentiment Analysis usuallyperformed by using several machine learning techniques. One of the main issues inSentiment Analysis is the availability of tagged resources to properly train super-vised classification algorithms. This is of particular concern in languages other thanEnglish, such as Spanish, where scarcity of these resources is the norm. In addition,most basic emotion datasets available in Spanish are rather small, containing a few hundred (or thousand) samples. Usually, the samples only contain a short text(frequently a comment) and a tag (the basic emotion), omitting crucial contextualinformation that may help to improve the classification task results. In this paper, theimpact of using contextual information is measured on a recently published Spanishbasic emotion dataset and the baseline architecture proposed in the Semantic Eval-uation 2019 competition. This particular dataset has two main advantages for thispaper. First, it was compiled using Distant Supervision and as a result it containsseveral hundred thousand samples. Secondly, the authors included valuable contex-tual information for each comment. The results show that contextual information,such as news headlines or summaries, helps improve the classification accuracy overa dataset of distantly supervised basic emotion labelled comments. Fil: Tessore, Juan Pablo. Universidad Nacional del Noroeste de la Provincia de Buenos Aires. Instituto de Investigación y Transferencia en Tecnología. Centro Asociado a la Comisión de Investigaciones Científicas de la Provincia de Buenos Aires; Argentina. Fil: Esnaola, Leonardo Martín. Universidad Nacional del Noroeste de la Provincia de Buenos Aires. Instituto de Investigación y Transferencia en Tecnología. Centro Asociado a la Comisión de Investigaciones Científicas de la Provincia de Buenos Aires; Argentina. Fil: Ramón, Hugo Dionisio. Universidad Nacional del Noroeste de la Provincia de Buenos Aires. Instituto de Investigación y Transferencia en Tecnología. Centro Asociado a la Comisión de Investigaciones Científicas de la Provincia de Buenos Aires; Argentina. Fil: Lanzarini, Laura. Universidad Nacional de La Plata. Facultad de Informática. Instituto de Investigación en Informática. Centro Asociado a la Comisión de Investigaciones Científicas de la Provincia de Buenos Aires; Argentina. Fil: Baldassarri, Sandra. Universidad de Zaragoza. Departamento de Informática e Ingeniería de Sistemas; España. Con referato 2023-12-07T18:20:25Z info:eu-repo/date/embargoEnd/2023-09-15 2023-12-07T18:20:25Z 2022-09-15 info:eu-repo/semantics/article info:ar-repo/semantics/artículo info:eu-repo/semantics/acceptedVersion info:eu-repo/semantics/article info:ar-repo/semantics/artículo info:eu-repo/semantics/acceptedVersion info:eu-repo/semantics/article info:ar-repo/semantics/artículo info:eu-repo/semantics/acceptedVersion Tessore, J., Esnaola L., Ramon, H., Lanzarini, L. y Baldassarri, S. (2022). Contextual information usage for the enhancement of basic emotion classification in a weakly labelled social network dataset in Spanish. Multimedia Tools and Applications, 82, 9871-9890. 1573-7721 1380-7501 http://repositorio.unnoba.edu.ar/xmlui/handle/23601/645 eng info:eu-repo/grantAgreement/UNNOBA/SIB2017/EXP 2101/2022/AR. Buenos Aires/Inteligencia artificial como herramienta para innovar y dinamizar procesos https://link.springer.com/article/10.1007/s11042-022-13750-x info:eu-repo/semantics/embargoedAccess https://creativecommons.org/licenses/by-nc-nd/2.5/ar/ application/pdf application/pdf Springer Science+Business Media, LLC, part of Springer Nature Multimedia Tools and Applications

Ejemplares similares