A Framework for Perception Analysis of Social Media Data During Disease Outbreaks: Uncovering Patterns of Resentment Towards Bats

Simple item page
dc.contributor.author Okpala, Izunna
dc.contributor.author Romera Rodriguez, Guillermo
dc.contributor.author Han, Chaeeun
dc.contributor.author Meierhofer, Melissa
dc.contributor.author Mammola, Stefano
dc.contributor.author Halse, Shane
dc.contributor.author Kropczynski, Jess
dc.contributor.author Johnson, Joseph
dc.date.accessioned 2023-12-26T18:38:37Z
dc.date.available 2023-12-26T18:38:37Z
dc.date.issued 2024-01-03
dc.identifier.isbn 978-0-9981331-7-1
dc.identifier.other a6ffd3f4-e9b6-4247-acd4-4d6eec07efc9
dc.identifier.uri https://hdl.handle.net/10125/106683
dc.language.iso eng
dc.relation.ispartof Proceedings of the 57th Hawaii International Conference on System Sciences
dc.rights Attribution-NonCommercial-NoDerivatives 4.0 International
dc.rights.uri https://creativecommons.org/licenses/by-nc-nd/4.0/
dc.subject Data Analytics, Data Mining, and Machine Learning for Social Media
dc.subject bats
dc.subject decision-making tools
dc.subject natural language processing
dc.subject perception analysis
dc.subject text analysis
dc.title A Framework for Perception Analysis of Social Media Data During Disease Outbreaks: Uncovering Patterns of Resentment Towards Bats
dc.type Conference Paper
dc.type.dcmi Text
dcterms.abstract Despite the growing number of natural language processing (NLP) tools developed for decision-makers to leverage social media for public perception evaluation during crises, a more robust framework is needed. This study explores a domain-specific machine learning framework for perception analysis using tweets about bats during disease outbreaks as a case study. Zoonotic disease outbreaks such as COVID-19 and Ebola are often attributed to bats and have resulted in unnecessary culling of wildlife; therefore, this is a case where perception is meaningful to a species. Analysis of 15,968 tweets showed a pattern in which tweets with anti-bat perceptions were most common during the early phases of an outbreak but declined over time while remaining negative, with 87.6% reliability of the framework according to manual coding of 300 randomly selected tweets. The framework can help stakeholders understand trends in public perception in near real-time and guide responses to spreading misinformation.
dcterms.extent 10 pages
prism.startingpage 2475
Files
Original bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
0241.pdf
Size:
951.86 KB
Format:
Adobe Portable Document Format
Description:
Download
Collections
Data Analytics, Data Mining, and Machine Learning for Social Media