COVID-19 Public Opinion: A Twitter Healthcare Data Processing Using Machine Learning Methodologies.
Int J Environ Res Public Health
; 20(1)2022 12 27.
Article
en En
| MEDLINE
| ID: mdl-36612755
ABSTRACT
The COVID-19 pandemic has shattered the whole world, and due to this, millions of people have posted their sentiments toward the pandemic on different social media platforms. This resulted in a huge information flow on social media and attracted many research studies aimed at extracting useful information to understand the sentiments. This paper analyses data imported from the Twitter API for the healthcare sector, emphasizing sub-domains, such as vaccines, post-COVID-19 health issues and healthcare service providers. The main objective of this research is to analyze machine learning models for classifying the sentiments of people and analyzing the direction of polarity by considering the views of the majority of people. The inferences drawn from this analysis may be useful for concerned authorities as they work to make appropriate policy decisions and strategic decisions. Various machine learning models were developed to extract the actual emotions, and results show that the support vector machine model outperforms with an average accuracy of 82.67% compared with the logistic regression, random forest, multinomial naïve Bayes and long short-term memory models, which present 78%, 77%, 68.67% and 75% accuracy, respectively.
Palabras clave
Texto completo:
1
Bases de datos:
MEDLINE
Asunto principal:
Medios de Comunicación Sociales
/
COVID-19
Tipo de estudio:
Prognostic_studies
Límite:
Humans
Idioma:
En
Revista:
Int J Environ Res Public Health
Año:
2022
Tipo del documento:
Article
País de afiliación:
India