TABLE 1.

Testing performance of machine learning classifiers for Salmonella prevalence prediction

Model name	Optimal hyperparameters	Precision	Recall	F1 score	Accuracy
Adaptive Boosting Classifier	learning_rate: 0.1, n_estimators: 100	0.90	0.87	0.87	0.87
Decision Tree Classifier	criterion: “gini,” max_depth: None, min_samples_leaf: 2, min_samples_split: 5	0.96	0.96	0.96	0.96
Gaussian Naive Bayes	var_smoothing: 1 × 10–9	0.87	0.87	0.87	0.87
Logistic Regression	C: 100, solver: “newton-cg”	0.84	0.83	0.83	0.83
Multi-layer Perceptron Classifier	activation: “tanh,” alpha: 0.0001, hidden_layer_sizes: (100, 50, 50), learning_rate: “constant,” solver: “adam”	0.37	0.61	0.46	0.61
Random Forest Classifier	max_depth: None, max_features: “log2”, min_samples_split: 10, n_estimators: 100	0.87	0.87	0.87	0.87
Stochastic Gradient Descent Classifier	alpha: 0.001, learning_rate: “optimal,” loss: “perceptron,” penalty: “l1”	0.37	0.61	0.46	0.61