Attention-Based Bi-DLSTM for Sentiment Analysis of Beijing Opera Lyrics

Table 1

Model experiment comparison.

Learning model	Precision	Recall	F1-score
CNN	0.625	0.619	0.628
CNN-LSTM	0.731	0.743	0.741
Bi-LSTM	0.763	0.757	0.759
Bi-LSTM+attention	0.804	0.798	0.810
Bi-DLSTM	0.823	0.833	0.824
Bi-DLSTM+attention	0.854	0.846	0.849