Buzz Tweet Classification Based on Text and Image Features of Tweets Using Multi-Task Learning

This study investigates social media trends and proposes a buzz tweet classification method to explore the factors causing the buzz phenomenon on Twitter. It is difficult to identify the causes of the buzz phenomenon based solely on texts posted on Twitter. It is expected that by limiting the tweets...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Reishi Amitani, Kazuyuki Matsumoto, Minoru Yoshida, Kenji Kita
Formato: article
Lenguaje:EN
Publicado: MDPI AG 2021
Materias:
T
Acceso en línea:https://doaj.org/article/a4e48647ded1406aadcc5f41e1531b2f
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:a4e48647ded1406aadcc5f41e1531b2f
record_format dspace
spelling oai:doaj.org-article:a4e48647ded1406aadcc5f41e1531b2f2021-11-25T16:31:46ZBuzz Tweet Classification Based on Text and Image Features of Tweets Using Multi-Task Learning10.3390/app1122105672076-3417https://doaj.org/article/a4e48647ded1406aadcc5f41e1531b2f2021-11-01T00:00:00Zhttps://www.mdpi.com/2076-3417/11/22/10567https://doaj.org/toc/2076-3417This study investigates social media trends and proposes a buzz tweet classification method to explore the factors causing the buzz phenomenon on Twitter. It is difficult to identify the causes of the buzz phenomenon based solely on texts posted on Twitter. It is expected that by limiting the tweets to those with attached images and using the characteristics of the images and the relationships between the text and images, a more detailed analysis than that of with text-only tweets can be conducted. Therefore, an analysis method was devised based on a multi-task neural network that uses both the features extracted from the image and text as input and the buzz class (buzz/non-buzz) and the number of “likes (favorites)” and “retweets (RTs)” as output. The predictions made using a single feature of the text and image were compared with the predictions using a combination of multiple features. The differences between buzz and non-buzz features were analyzed based on the cosine similarity between the text and the image. The buzz class was correctly identified with a correctness rate of approximately 80% for all combinations of image and text features, with the combination of BERT and VGG16 providing the highest correctness rate.Reishi AmitaniKazuyuki MatsumotoMinoru YoshidaKenji KitaMDPI AGarticlemulti-task learningbuzz classificationsocial mediatrend analysisTechnologyTEngineering (General). Civil engineering (General)TA1-2040Biology (General)QH301-705.5PhysicsQC1-999ChemistryQD1-999ENApplied Sciences, Vol 11, Iss 10567, p 10567 (2021)
institution DOAJ
collection DOAJ
language EN
topic multi-task learning
buzz classification
social media
trend analysis
Technology
T
Engineering (General). Civil engineering (General)
TA1-2040
Biology (General)
QH301-705.5
Physics
QC1-999
Chemistry
QD1-999
spellingShingle multi-task learning
buzz classification
social media
trend analysis
Technology
T
Engineering (General). Civil engineering (General)
TA1-2040
Biology (General)
QH301-705.5
Physics
QC1-999
Chemistry
QD1-999
Reishi Amitani
Kazuyuki Matsumoto
Minoru Yoshida
Kenji Kita
Buzz Tweet Classification Based on Text and Image Features of Tweets Using Multi-Task Learning
description This study investigates social media trends and proposes a buzz tweet classification method to explore the factors causing the buzz phenomenon on Twitter. It is difficult to identify the causes of the buzz phenomenon based solely on texts posted on Twitter. It is expected that by limiting the tweets to those with attached images and using the characteristics of the images and the relationships between the text and images, a more detailed analysis than that of with text-only tweets can be conducted. Therefore, an analysis method was devised based on a multi-task neural network that uses both the features extracted from the image and text as input and the buzz class (buzz/non-buzz) and the number of “likes (favorites)” and “retweets (RTs)” as output. The predictions made using a single feature of the text and image were compared with the predictions using a combination of multiple features. The differences between buzz and non-buzz features were analyzed based on the cosine similarity between the text and the image. The buzz class was correctly identified with a correctness rate of approximately 80% for all combinations of image and text features, with the combination of BERT and VGG16 providing the highest correctness rate.
format article
author Reishi Amitani
Kazuyuki Matsumoto
Minoru Yoshida
Kenji Kita
author_facet Reishi Amitani
Kazuyuki Matsumoto
Minoru Yoshida
Kenji Kita
author_sort Reishi Amitani
title Buzz Tweet Classification Based on Text and Image Features of Tweets Using Multi-Task Learning
title_short Buzz Tweet Classification Based on Text and Image Features of Tweets Using Multi-Task Learning
title_full Buzz Tweet Classification Based on Text and Image Features of Tweets Using Multi-Task Learning
title_fullStr Buzz Tweet Classification Based on Text and Image Features of Tweets Using Multi-Task Learning
title_full_unstemmed Buzz Tweet Classification Based on Text and Image Features of Tweets Using Multi-Task Learning
title_sort buzz tweet classification based on text and image features of tweets using multi-task learning
publisher MDPI AG
publishDate 2021
url https://doaj.org/article/a4e48647ded1406aadcc5f41e1531b2f
work_keys_str_mv AT reishiamitani buzztweetclassificationbasedontextandimagefeaturesoftweetsusingmultitasklearning
AT kazuyukimatsumoto buzztweetclassificationbasedontextandimagefeaturesoftweetsusingmultitasklearning
AT minoruyoshida buzztweetclassificationbasedontextandimagefeaturesoftweetsusingmultitasklearning
AT kenjikita buzztweetclassificationbasedontextandimagefeaturesoftweetsusingmultitasklearning
_version_ 1718413156680728576