An Efficient Sentiment Classification Method with the Help of Neighbors and a Hybrid of RNN Models

Table 5

Performance comparison of different models.

Model	Metric	Dataset
		BG	BIS	MVSA-Single	Twitter	IMDB
CNN	CNN	Accuracy	98.30	97.02	80.86	94.83
		Precision	98.23	98.21	81.09	94.82
		Recall	98.35	96.24	80.86	94.83
		F1-score	98.30	97.22	80.92	94.82
GRU	GRU	Accuracy	98.34	97.06	80.90	94.89
		Precision	98.30	98.24	81.03	94.89
		Recall	98.37	96.27	80.80	94.90
		F1-score	98.32	97.24	80.86	94.88
LSTM	LSTM	Accuracy	98.40	97.12	80.96	94.93
		Precision	98.33	98.31	81.19	94.92
		Recall	98.45	96.34	80.96	94.93
		F1-score	98.40	97.32	81.02	94.92
LSTM-CNN	LSTM-CNN	Accuracy	98.85	97.57	81.41	95.38
		Precision	98.78	98.76	81.64	95.37
		Recall	98.90	96.79	81.41	95.38
		F1-score	98.85	97.77	81.47	95.37
Bi-LSTM	Bi-LSTM	Accuracy	99.10	97.82	81.66	95.63
		Precision	99.03	99.01	81.89	95.62
		Recall	99.15	97.04	81.66	95.63
		F1-score	99.10	98.02	81.72	95.62
Bi-GRU	Bi-GRU	Accuracy	99.20	97.92	81.76	95.73
		Precision	99.13	99.11	81.99	95.72
		Recall	99.25	97.14	81.76	95.73
		F1-score	99.20	98.12	81.82	95.72
DTSC	DTSC	Accuracy	99.60	98.32	82.19	96.13
		Precision	99.55	99.51	82.39	96.12
		Recall	99.65	97.54	82.16	96.13
		F1-score	99.60	98.52	82.22	96.12