Optimizing machine learning models for predicting anemia among under-five children in Ethiopia: insights from Ethiopian demographic and health survey data

Table 4 Metrics of model accuracy for each model as assessed using test data

	Machine learning algorithms
	Decision Tree	Random Forest	Naïve Bayes	Logistic Regression	KNN	SVM
	%(CI)	%(CI)	%(CI)	%(CI)	%(CI)	%(CI)
Accuracy	68.40(65.6–72.3)	81.16(77.1–85.4)	53.06(49.1–57.2)	54.79(50.1–58.3)	69.96(65.9–74.1)	59.94(55.2–53.6)
Sensitivity	70.21	83.07	21.90	44.68	58.50	59.70
Specificity	66.58	79.26	84.22	64.89	70.90	64.60
PP Value	67.75	80.02	58.12	56.00	67.30	62.80
NP Value	69.09	82.40	51.88	53.98	75.00	68.6
AUC	56.30	81.80	54.4	56.10	72.90	61.80

ISSN: 1471-2431