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

Table 4

The results of model experiments (microblogPCU).

Model	Precision	Recall	F1
AB + LR	0.865	0.821	0.842
AB + NB	0.817	0.826	0.821
AB + SVM	0.861	0.842	0.850
W2V + Bi-LSTM	0.850	0.853	0.851
AB + Bi-LSTM	0.899	0.870	0.884
AB + SA Bi-LSTM	0.912	0.891	0.901