Table 4:

Machine learning algorithms, hyper-parameters and performance using 13 handcrafted features. We perform a 3-fold cross validation to evaluate the performance of the model. F1 scores are reported.

Algorithm	Hyper- parameter	Values	F1
KNN	n_neighbors	5, 6, 7, 8, 9, 10	0.65
Linear SVC	gamma	0.001, 0.01, 0.1, 1	0.71
	C	0.001, 0.01, 0.1, 1, 10, 100
SVC (RBF kernel)	gamma	0.001, 0.01, 0.1, 1	0.68
	C	0.001, 0.01, 0.1, 1, 10, 100
Logistic Regression (l2)	C	np.logspace(−4, 4, 20)	0.73
Random Forest	max_depth	3, 5, 7	0.56
	n_estimators	range(10,101,10)
	min_samples_split	3, 5
Gradient Boosting Tree	max_depth	3, 4, 5	0.68
	subsample	0.5, 0.6, 0.8, 0.85, 0.9, 0.95, 1.0
	learning_rate	0.01, 0.03, 0.05, 0.075, 0.1, 0.2
	n_estimators	range(10,101,10)
Neural Network	hidden layers	1, 2, 3	0.71
	hidden units	100, 200, 300
	dropout rate	0, 0.25, 0.5