COVID-19 vaccination helps protect people from getting the virus. Some people show up normal signs from the vaccine, which indicates that their body is building protection. However, adverse effects on people could cause long-term health problems. Severe allergic reactions, Myocarditis, and Pericarditis appeared in the vaccinated people that have been reported to the (FDA/CDC) Vaccine Adverse Event Reporting System (VAERS). In fact, other possible effects are still being studied in clinical trials. In the present work, the adverse reactions caused by Covid-19 vaccines of Pfizer/BioNTech, Moderna, and JJ Johnson & Johnson manufacturers are studied. Specifically, the supervised machine learning approach is utilized to discriminate body reactions against the vaccine and provide a decision-making model for the vaccine recipients. The model study and analyze the recipients’ reactions whether they showed mild, moderate, or severe acute syndromes to reduce the fatality rates. To validate our model, a dataset of more than 52k records with 18 informative attributes provided by VAERS has been utilized, and three supervised learning algorithms have been implemented in Python which are Decision Tree, Support Vector Machine, and Naïve Bayes to conduct two experiments. A simple splitting percentage method was performed in the first one, while a k-Folds Cross-validation technique was used in the second experiment with k=5. The model showed a promising result with stable performance in both experiments, the Decision Tree outperformed other algorithms with a predictive rate of 0.91999 in the first experiment, and 0.91369 in the second one.
Albayati, Mohammed Basil and Altamimi, Ahmad Mousa
"Analyzing COVID-19 Vaccine Adverse Reactions Using Machine Learning Techniques,"
Karbala International Journal of Modern Science: Vol. 9
, Article 11.
Available at: https://doi.org/10.33640/2405-609X.3271
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.