Crowdsourcing and machine learning approaches for extracting entities indicating potential foodborne outbreaks from social media

Abstract Foodborne outbreaks are a serious but preventable threat to public health that often lead to illness, loss of life, significant economic loss, and the erosion of consumer confidence. Understanding how consumers respond when interacting with foods, as well as extracting information from post...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Dandan Tao, Dongyu Zhang, Ruofan Hu, Elke Rundensteiner, Hao Feng
Formato: article
Lenguaje:EN
Publicado: Nature Portfolio 2021
Materias:
R
Q
Acceso en línea:https://doaj.org/article/687c6d57923345e88d848bcfb42e1505
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:687c6d57923345e88d848bcfb42e1505
record_format dspace
spelling oai:doaj.org-article:687c6d57923345e88d848bcfb42e15052021-11-08T10:50:50ZCrowdsourcing and machine learning approaches for extracting entities indicating potential foodborne outbreaks from social media10.1038/s41598-021-00766-w2045-2322https://doaj.org/article/687c6d57923345e88d848bcfb42e15052021-11-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-00766-whttps://doaj.org/toc/2045-2322Abstract Foodborne outbreaks are a serious but preventable threat to public health that often lead to illness, loss of life, significant economic loss, and the erosion of consumer confidence. Understanding how consumers respond when interacting with foods, as well as extracting information from posts on social media may provide new means of reducing the risks and curtailing the outbreaks. In recent years, Twitter has been employed as a new tool for identifying unreported foodborne illnesses. However, there is a huge gap between the identification of sporadic illnesses and the early detection of a potential outbreak. In this work, the dual-task BERTweet model was developed to identify unreported foodborne illnesses and extract foodborne-illness-related entities from Twitter. Unlike previous methods, our model leveraged the mutually beneficial relationships between the two tasks. The results showed that the F1-score of relevance prediction was 0.87, and the F1-score of entity extraction was 0.61. Key elements such as time, location, and food detected from sentences indicating foodborne illnesses were used to analyze potential foodborne outbreaks in massive historical tweets. A case study on tweets indicating foodborne illnesses showed that the discovered trend is consistent with the true outbreaks that occurred during the same period.Dandan TaoDongyu ZhangRuofan HuElke RundensteinerHao FengNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-12 (2021)
institution DOAJ
collection DOAJ
language EN
topic Medicine
R
Science
Q
spellingShingle Medicine
R
Science
Q
Dandan Tao
Dongyu Zhang
Ruofan Hu
Elke Rundensteiner
Hao Feng
Crowdsourcing and machine learning approaches for extracting entities indicating potential foodborne outbreaks from social media
description Abstract Foodborne outbreaks are a serious but preventable threat to public health that often lead to illness, loss of life, significant economic loss, and the erosion of consumer confidence. Understanding how consumers respond when interacting with foods, as well as extracting information from posts on social media may provide new means of reducing the risks and curtailing the outbreaks. In recent years, Twitter has been employed as a new tool for identifying unreported foodborne illnesses. However, there is a huge gap between the identification of sporadic illnesses and the early detection of a potential outbreak. In this work, the dual-task BERTweet model was developed to identify unreported foodborne illnesses and extract foodborne-illness-related entities from Twitter. Unlike previous methods, our model leveraged the mutually beneficial relationships between the two tasks. The results showed that the F1-score of relevance prediction was 0.87, and the F1-score of entity extraction was 0.61. Key elements such as time, location, and food detected from sentences indicating foodborne illnesses were used to analyze potential foodborne outbreaks in massive historical tweets. A case study on tweets indicating foodborne illnesses showed that the discovered trend is consistent with the true outbreaks that occurred during the same period.
format article
author Dandan Tao
Dongyu Zhang
Ruofan Hu
Elke Rundensteiner
Hao Feng
author_facet Dandan Tao
Dongyu Zhang
Ruofan Hu
Elke Rundensteiner
Hao Feng
author_sort Dandan Tao
title Crowdsourcing and machine learning approaches for extracting entities indicating potential foodborne outbreaks from social media
title_short Crowdsourcing and machine learning approaches for extracting entities indicating potential foodborne outbreaks from social media
title_full Crowdsourcing and machine learning approaches for extracting entities indicating potential foodborne outbreaks from social media
title_fullStr Crowdsourcing and machine learning approaches for extracting entities indicating potential foodborne outbreaks from social media
title_full_unstemmed Crowdsourcing and machine learning approaches for extracting entities indicating potential foodborne outbreaks from social media
title_sort crowdsourcing and machine learning approaches for extracting entities indicating potential foodborne outbreaks from social media
publisher Nature Portfolio
publishDate 2021
url https://doaj.org/article/687c6d57923345e88d848bcfb42e1505
work_keys_str_mv AT dandantao crowdsourcingandmachinelearningapproachesforextractingentitiesindicatingpotentialfoodborneoutbreaksfromsocialmedia
AT dongyuzhang crowdsourcingandmachinelearningapproachesforextractingentitiesindicatingpotentialfoodborneoutbreaksfromsocialmedia
AT ruofanhu crowdsourcingandmachinelearningapproachesforextractingentitiesindicatingpotentialfoodborneoutbreaksfromsocialmedia
AT elkerundensteiner crowdsourcingandmachinelearningapproachesforextractingentitiesindicatingpotentialfoodborneoutbreaksfromsocialmedia
AT haofeng crowdsourcingandmachinelearningapproachesforextractingentitiesindicatingpotentialfoodborneoutbreaksfromsocialmedia
_version_ 1718442624056033280