Prediction of post-covid-19 using supervised machine learning techniques

Abstract

The COVID-19 pandemic has had a profound impact on global health, necessitating the development of predictive models to manage and mitigate its effects. Early diagnosis is crucial for preventing the progression of diseases that can significantly endanger human life. This study explores the application of supervised machine learning techniques to predict Post-COVID-19 outcomes, including long-term health complications and recovery trajectories. In this study, we utilized 10 advanced supervised machine learning algorithms, including both stand-alone models (Decision Tree, Random Forest, Logistic Regression, K-Nearest Neighbors, Support Vector Machine, and Gaussian Naive Bayes) and ensemble learning techniques (Bagging Decision Tree Ensemble, Boosting Decision Tree Ensemble, Voting Ensemble, and Stacked Generalization – Stacking Ensemble). These models were applied to analyze and predict the presence of COVID-19 using the COVID-19 Symptoms and Presence dataset from Kaggle. The performance of each model was evaluated using an 80:20 train-test split as well as 5, 10, 15, 20, and 25-fold cross-validation. Evaluation metrics included accuracy, precision, recall, F1-score, and the confusion matrix. The results indicate that the Decision Tree algorithm outperformed the other models, achieving an accuracy of 98.81%, a precision of 1.00, a recall of 0.98, and an F1-score of 0.99. Our results indicate that machine learning models can effectively predict Post-COVID-19 conditions, providing valuable insights for healthcare providers and policymakers.