Table 5.
Embedding method | Classification | Accuracy | Precision | Recall | F1 |
---|---|---|---|---|---|
word2vec | LR | 0.824 | 0.834 | 0.830 | 0.832 |
LDA | 0.825 | 0.835 | 0.831 | 0.833 | |
KNN | 0.825 | 0.828 | 0.841 | 0.834 | |
CART | 0.812 | 0.834 | 0.800 | 0.817 | |
NB | 0.688 | 0.665 | 0.815 | 0.732 | |
SVM | 0.838 | 0.843 | 0.849 | 0.846 | |
XGBoost | 0.840 | 0.833 | 0.870 | 0.851 | |
RDForest | 0.838 | 0.824 | 0.881 | 0.851 | |
Ising-word2vec | LR | 0.818 | 0.825 | 0.829 | 0.826 |
LDA | 0.811 | 0.825 | 0.812 | 0.818 | |
KNN | 0.831 | 0.836 | 0.845 | 0.840 | |
CART | 0.809 | 0.835 | 0.794 | 0.813 | |
NB | 0.681 | 0.660 | 0.807 | 0.726 | |
SVM | 0.836 | 0.842 | 0.847 | 0.844 | |
XGBoost | 0.837 | 0.831 | 0.867 | 0.848 | |
RDForest | 0.841 | 0.828 | 0.881 | 0.853 |