Table 1 Hyperparameter settings

KNN

Number of neighbors (13); type of algorithm (auto); leaf size (30); power parameter p (1);

SVM

C (10); Kernel (rbf); gamma (0.1);

Decision tree

Maximum depth (20); minimum samples leaf (1);

Random forest

Number of estimators (300); maximum features (auto); maximum depth (70); minimum samples split (5) minimum samples leaf (4); bootstrap (true);

XGBoost

Number of estimators (500); learning rate (0.01); maximum depth (10); minimum child weight (5);

Neural network (CNN) + Word2vec

Batch size (64); number of epochs (100); optimizer (Adam); dropout (0.3); number of hidden layers (2); filter size (1); number of filters (200); minimum word frequency (5); iterations (100); layer size (300); window size (25);