Table 3.

Optimal hyper-parameters and Python libraries used in this study

Classifier	Python library	Parameter
DT	DecisionTreeClassifier()	criterion = Gini; min_samples_leaf = 10
RF	RandomForestClassifier()	criterion = Gini; n_estimators = 100
KNN	KNeighborsClassifier()	k = 5; weights = uniform
ANN	tf.keras.models.Sequential()	2 hidden layers with hidden_layer_sizes = 80, 40; activation = ReLU (hidden layers), sigmoid (output layer); epochs = 30; batch_size = 10
SVM	svm.SVC()	kernel = RBF
LR	LogisticRegression()	solver = liblinear