A Graph-Based Method for Predicting the Helpfulness of Product Opinions

This paper presents a new approach to predict the helpfulness of opinions. Usually, researchers in this area use tables of attribute-value to aggregate the features that represent the evaluated texts. Although that representation is common, it considers that the objects are independent. We argue tha...

Descripción completa

Detalles Bibliográficos
Autores: de Sousa, Rogério Figueredo, Anchiêta, Rafael Tôrres, Nunes, Maria das Graças Volpe
Tipo de recurso: artículo
Estado:Versión publicada
Fecha de publicación:2020
País:Brasil
Institución:Sociedade Brasileira de Computação (SBC)
Repositorio:Brazilian Journal of Information Systems
Idioma:inglés
OAI Identifier:oai:journals-sol.sbc.org.br:article/821
Acceso en línea:https://journals-sol.sbc.org.br/index.php/isys/article/view/821
Access Level:acceso abierto
Palabra clave:Natural Language Processing
Helpfulness Prediction
Opinion Mining
id BR_ffe9391c3d102c600fa9decbca50e794
oai_identifier_str oai:journals-sol.sbc.org.br:article/821
network_acronym_str BR
network_name_str Brasil
repository_id_str
spelling A Graph-Based Method for Predicting the Helpfulness of Product OpinionsNatural Language ProcessingHelpfulness PredictionOpinion MiningThis paper presents a new approach to predict the helpfulness of opinions. Usually, researchers in this area use tables of attribute-value to aggregate the features that represent the evaluated texts. Although that representation is common, it considers that the objects are independent. We argue that among the discriminant factors of the helpfulness of opinions, there are dependent factors of the relationship among the opinion-forming elements. Thus, we modeled this task as a network, considering the information of relations among objects in the network (comments, stars, and words). A regularization technique of graphs is used to extract the relevant features of graph structure and, after that, the comments are classified as helpful or unhelpful. We compared our network model with two baselines methods, one based on fuzzy logic and another based on Neural Networks. Our model outperformed the fuzzy logic and Neutal Network methods in 0.17 and 0.19 of F-measure, respectively. The main advantages of our approach are that few data are necessary to helpfulness classification and the relationships may help in the understanding the classification, explaining the reasons for a determinate classification.Sociedade Brasileira de Computação2020-07-29info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionapplication/pdfhttps://journals-sol.sbc.org.br/index.php/isys/article/view/82110.5753/isys.2020.821iSys - Revista Brasileira de Sistemas de Informação; v. 13 n. 4 (2020); 06-21iSys - Brazilian Journal of Information Systems; Vol. 13 No. 4 (2020); 06-211984-290210.5753/isys.2020.4reponame:Brazilian Journal of Information Systemsinstname:Sociedade Brasileira de Computação (SBC)instacron:SBCenghttps://journals-sol.sbc.org.br/index.php/isys/article/view/821/749Copyright (c) 2020 iSys - Brazilian Journal of Information Systemsinfo:eu-repo/semantics/openAccessde Sousa, Rogério FigueredoAnchiêta, Rafael TôrresNunes, Maria das Graças Volpe2020-07-31T19:03:58Zoai:journals-sol.sbc.org.br:article/821Revistahttps://journals-sol.sbc.org.br/index.php/isys/ONGhttps://journals-sol.sbc.org.br/index.php/isys/oaipublicacoes@sbc.org.br1984-29021984-2902opendoar:2020-07-31T19:03:58Brazilian Journal of Information Systems - Sociedade Brasileira de Computação (SBC)false
dc.title.none.fl_str_mv A Graph-Based Method for Predicting the Helpfulness of Product Opinions
title A Graph-Based Method for Predicting the Helpfulness of Product Opinions
spellingShingle A Graph-Based Method for Predicting the Helpfulness of Product Opinions
de Sousa, Rogério Figueredo
Natural Language Processing
Helpfulness Prediction
Opinion Mining
title_short A Graph-Based Method for Predicting the Helpfulness of Product Opinions
title_full A Graph-Based Method for Predicting the Helpfulness of Product Opinions
title_fullStr A Graph-Based Method for Predicting the Helpfulness of Product Opinions
title_full_unstemmed A Graph-Based Method for Predicting the Helpfulness of Product Opinions
title_sort A Graph-Based Method for Predicting the Helpfulness of Product Opinions
dc.creator.none.fl_str_mv de Sousa, Rogério Figueredo
Anchiêta, Rafael Tôrres
Nunes, Maria das Graças Volpe
author de Sousa, Rogério Figueredo
author_facet de Sousa, Rogério Figueredo
Anchiêta, Rafael Tôrres
Nunes, Maria das Graças Volpe
author_role author
author2 Anchiêta, Rafael Tôrres
Nunes, Maria das Graças Volpe
author2_role author
author
dc.subject.por.fl_str_mv Natural Language Processing
Helpfulness Prediction
Opinion Mining
topic Natural Language Processing
Helpfulness Prediction
Opinion Mining
description This paper presents a new approach to predict the helpfulness of opinions. Usually, researchers in this area use tables of attribute-value to aggregate the features that represent the evaluated texts. Although that representation is common, it considers that the objects are independent. We argue that among the discriminant factors of the helpfulness of opinions, there are dependent factors of the relationship among the opinion-forming elements. Thus, we modeled this task as a network, considering the information of relations among objects in the network (comments, stars, and words). A regularization technique of graphs is used to extract the relevant features of graph structure and, after that, the comments are classified as helpful or unhelpful. We compared our network model with two baselines methods, one based on fuzzy logic and another based on Neural Networks. Our model outperformed the fuzzy logic and Neutal Network methods in 0.17 and 0.19 of F-measure, respectively. The main advantages of our approach are that few data are necessary to helpfulness classification and the relationships may help in the understanding the classification, explaining the reasons for a determinate classification.
publishDate 2020
dc.date.none.fl_str_mv 2020-07-29
dc.type.driver.fl_str_mv info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion
format article
status_str publishedVersion
dc.identifier.uri.fl_str_mv https://journals-sol.sbc.org.br/index.php/isys/article/view/821
10.5753/isys.2020.821
url https://journals-sol.sbc.org.br/index.php/isys/article/view/821
identifier_str_mv 10.5753/isys.2020.821
dc.language.iso.fl_str_mv eng
language eng
dc.relation.none.fl_str_mv https://journals-sol.sbc.org.br/index.php/isys/article/view/821/749
dc.rights.driver.fl_str_mv Copyright (c) 2020 iSys - Brazilian Journal of Information Systems
info:eu-repo/semantics/openAccess
rights_invalid_str_mv Copyright (c) 2020 iSys - Brazilian Journal of Information Systems
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.publisher.none.fl_str_mv Sociedade Brasileira de Computação
publisher.none.fl_str_mv Sociedade Brasileira de Computação
dc.source.none.fl_str_mv iSys - Revista Brasileira de Sistemas de Informação; v. 13 n. 4 (2020); 06-21
iSys - Brazilian Journal of Information Systems; Vol. 13 No. 4 (2020); 06-21
1984-2902
10.5753/isys.2020.4
reponame:Brazilian Journal of Information Systems
instname:Sociedade Brasileira de Computação (SBC)
instacron:SBC
instname_str Sociedade Brasileira de Computação (SBC)
instacron_str SBC
institution SBC
reponame_str Brazilian Journal of Information Systems
collection Brazilian Journal of Information Systems
repository.name.fl_str_mv Brazilian Journal of Information Systems - Sociedade Brasileira de Computação (SBC)
repository.mail.fl_str_mv publicacoes@sbc.org.br
_version_ 1853662444560318464
score 15.301629

Similares en LA Referencia