Early Warning Scheme of COVID-19 related Internet Public Opinion based on RVM-L Model

Abstract

Internet public opinion is affected by many factors corresponding to insufficient data in the very short period, especially for emergency events related to the outbreak of coronavirus disease 2019 (COVID-19). To effectively support real-time analysis and accurate prediction, this paper proposes an early warning scheme, which comprehensively considers the multiple factors of Internet public opinion and the dynamic characteristics of burst events. A hybrid relevance vector machine and logistic regression (RVM-L) model is proposed that incorporates multivariate analysis, which adopts Lagrange interpolation to fill in the gaps and improve the forecasting effect based on insufficient data for COVID-19-related events. In addition, a novel metric critical interval is introduced to improve the early warning performance. Detailed experiments show that compared with existing schemes, the proposed RVM-L-based early warning scheme can achieve the prediction accuracy up to 96%, and the intervention within the critical interval can reduce the number of public opinions by 60%.

Internet Rumors During the COVID-19 Pandemic: Dynamics of Topics and Public Psychologies

“…Monitoring public opinions is an essential subject for modern society, especially for widely circulated rumors, which need the extra attention of governments to eliminate public misunderstanding and prevent further social conflicts in a timely manner ( 1 ). The outbreak of the COVID-19 pandemic led to a number of Internet rumors, which were characterized by a large number of high density, rapidly spreading retweets and widespread impact ( 20 – 23 ).…”

“…The endless rumors have brought more anxiety and distress in society than just the virus, and have inflicted great damage on people's enthusiasm to fight the pandemic ( 24 ). Therefore, it is essential to conduct targeted research into the Internet rumors that have been generated during this period ( 1 , 25 ), which will facilitate a better understanding of the real trends of COVID-19, avoid unnecessary anxiety, and restore normal social order and stability of society ( 26 ).…”

“…In relation to the propagation characteristics of pandemic rumors, the research of Singh et al ( 14 ) showed that COVID-19-informed users tended to use more narratives than misinformed users. Zhu et al ( 1 ) proposed an early warning scheme considering the multiple factors of Internet public opinion and the dynamic characteristics of burst events. Zeng et al ( 5 ) proposed an approach for fake news detection by comprehensively mining the semantic correlations between text content and attached images.…”

“…The monitoring and analysis of public opinions is an important task for modern society, as they may provide critical information that enables the timely capture of unanticipated public hotspots and trends, thus contributing to government business continuity ( 1 ). Along with the global ravages of the new coronavirus during the COVID-19 pandemic ( 2 , 3 ), the resulting information explosion poses an unprecedented challenge to public governance ( 4 ).…”

“…As the COVID-19 pandemic has gone through approximately one annual cycle, it is informative to conduct a comprehensive analysis on widespread Internet rumors. Such rumors have been spread around since the outbreak of the pandemic, and to clarify the characteristics of the rumors at different stages during the outbreak and circulation of the pandemic would help and provide understanding of public opinion responses to major public health emergencies ( 1 ).…”