Predicting COVID-19 vaccinators based on machine learning and sentiment analysis

Hadab Khalid Obayes, Khaldoon Hasan Alhussayni, Saba Mohammed Hussain

Abstract


In the past two years, the world witnessed the spread of the coronavirus (COVID-19) pandemic that disrupted the entire world, the only solution to this epidemic was health isolation, and with it everything stopped. When announcing the availability of a vaccine, the world was divided over the effectiveness and harms of this vaccine. This article provides an analysis of vaccinators and analysis of people's opinions of the vaccine's efficacy and whether negative or positive. Then a model is built to predict the future numbers of vaccinators and a model that predicts the number of negative opinions or tweets. The model consists of three stages: first, converting data sets into a synchronized time series, that is, the same place and time for vaccination and tweets. The second stage is building a prediction model and the third stage was descripting analysis of the prediction results. The autoregressive integrated moving averages (ARIMA) method was used after decomposing the components of ARIMA and choosing the optimal model, the best results obtained from seasonal ARIMA (SARIMA) for both predictions, the last stage is the descriptive analysis of the results and linking them together to obtain an analysis describing the change in the number of vaccinators and the number of negative tweets.


Keywords


COVID-19; Machine learning; Predicting; Sentiment analysis; Vaccine

Full Text:

PDF


DOI: https://doi.org/10.11591/eei.v12i3.4278

Refbacks

  • There are currently no refbacks.


Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

Bulletin of EEI Stats