Table 2.

Results of different methods, where “CNN” indicates the performance of the proposed model.

Method	Training set				Testing set
	Precision	Recall	F1-score	Accuracy	Precision	Recall	F1-score	Accuracy
SVM	0.96	0.95	0.96	0.9549	0.93	0.93	0.93	0.9315
MultinomialNB	0.93	0.92	0.92	0.9236	0.87	0.86	0.86	0.8600
LogisticRegression	0.93	0.93	0.93	0.9293	0.92	0.92	0.92	0.9175
KNeighborsClassifier	0.89	0.89	0.89	0.8911	0.90	0.89	0.89	0.8925
CNN	0.9947	0.9946	0.9946	0.9982	0.9594	0.9602	0.9596	0.9867