Textual Information Classification of Campus Network Public Opinion Based on BILSTM and ARIMA

Table 5

Comparison of model classification and prediction results.

	Negative			Neutral			Positive			Accuracy
	Precision	Recall		Precision	Recall		Precision	Recall
Modle1	0.4830	0.9924	0.6498	0.6429	0.0154	0.0301	0.7907	0.1478	0.2491	0.4909
Modle2	0.4845	0.9873	0.6500	0.6471	0.0189	0.0366	0.5545	0.1326	0.2140	0.4883
Modle3	0.4910	0.9762	0.6534	0.6296	0.0389	0.0731	0.4408	0.1457	0.2190	0.4928
Modle4	0.7746	0.8462	0.8088	0..8605	0.6770	0.7578	0.584	0.8439	0.6903	0.7718
Modle5	0.8176	0.8069	0.8122	0.8353	0.6988	0.7610	0.5724	0.948	0.7138	0.7762
Modle6	0.7870	0.83	0.8079	0.8678	0.6335	0.7323	0.5386	0.9682	0.6921	0.7602
Modle7	0.7470	0.7677	0.7572	0.6744	0.7883	0.7269	0.9355	0.2514	0.3964	0.7158
Modle8	0.9145	0.6915	0.7876	0.7134	0.9393	0.8109	0.957	0.7081	0.814	0.8021
Modle9	0.8894	0.7608	0.8201	0.7744	0.8763	0.8222	0.8110	0.8555	0.8326	0.8226