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

Table 3

The effect of the neighborhood technique.

Neighboring	Metrics	Dataset
		BG	BIS	MVSA-Single	Twitter	IMDB
Not taken into account	Accuracy	98.23	97.54	80.77	95.14	86.65
	Precision	98.46	98.29	80.20	95.18	89.45
	Recall	98.15	96.21	80.79	95.14	82.41
	F1-score	98.10	97.95	80.11	95.11	86.01
Taken into account	Accuracy	99.60	98.32	82.19	96.13	87.60
	Precision	99.55	99.51	82.39	96.12	91.06
	Recall	99.65	97.54	82.16	96.13	83.64
	F1-score	99.60	98.52	82.22	96.12	87.20

Bold values indicate that the proposed neighbourhood technique model has achieved better results.