<![CDATA[Asthma is a chronic respiratory disease that can significantly reduce a patient’s quality of life if it is not detected and treated properly. This study aims to compare the performance of Gaussian Naïve Bayes and Random Forest in predicting asthma using the Asthma Disease Dataset from Kaggle, which contains 2,392 patient records with an imbalanced class distribution: 2,268 non-asthma cases and 124 asthma cases. The steps taken include feature selection, preprocessing using StandardScaler, handling class imbalance with SMOTE applied exclusively to the training data, the classification process, and model evaluation using the metrics accuracy, precision, recall, F1-score, ROC AUC, Cohen’s Kappa, MCC, and 5-Fold Cross Validation. Test results showed that Random Forest achieved the highest accuracy of 0.904 with a precision of 0.080, while Gaussian Naïve Bayes produced a recall of 0.520, an F1-score of 0.134, and an ROC AUC of 0.638. These findings indicate that Random Forest is superior in terms of overall accuracy, while Gaussian Naïve Bayes is more effective in detecting asthma cases in the dataset used. The results of this study can serve as a reference in the development of decision support systems for asthma risk identification, although further validation using more diverse clinical data is still required.]]>
Copyrights © 2026
