Social Network Spam Detection Based on ALBERT and Combination of Bi-LSTM with Self-Attention

Table 3

The results of model experiments (weiboData).

Model	Precision	Recall	F1
AB + LR	0.868	0.829	0.848
AB + NB	0.810	0.814	0.811
AB + SVM	0.846	0.835	0.840
W2V + Bi-LSTM	0.844	0.841	0.842
AB + Bi-LSTM	0.889	0.862	0.875
AB + SA Bi-LSTM	0.903	0.863	0.882