Empowering child health: Harnessing machine learning to predict acute respiratory infections in Ethiopian under-fives using demographic and health survey insights

Table 4 Performance evaluation metrics of the trained machine learning models

Machine learning algorithms
Metrics	DT	RF	KNN	SVM	NB	LR	GB	XGB	LA	SVM + GB +XGB
Accuracy	0.79	0.78	0.76	0.81	0.74	0.80	0.81	0.81	0.79	0.86
Sensitivity (%)	74.3	79.4	62.8	78.2	52.5	74.3	76.9	85.8	73.1	84.6
Specificity (%)	83.3	76.9	88.4	83.3	96.1	85.8	85.8	75.6	85.8	89.7
Weighted F1-score	0.79	0.78	0.75	0.81	0.73	0.80	0.81	0.81	0.79	0.86
AUC-ROC	0.85	0.90	0.80	0.84	0.91	0.89	0.82	0.87	0.79	0.87
AUC-PRC	0.87	0.91	0.86	0.76	0.91	0.90	0.87	0.88	0.85	0.90

DT = Decision Tree, RF = Random Forest, KNN = K Nearest Neighbor, SVM = Support Vector Machines, NB = Naïve Bayes, LR = Logistic Regression, LA = Lasso Regression GB = Gradient Boosting, XGB = Extreme Gradient Boosting, SVM + GB + XGB = Ensemble model

ISSN: 1471-2334