Deep Learning-Based Fake Information Detection and Influence Evaluation

Table 3

The performance comparisons of the proposed method, the EANN method, and the SpotFake method.

Dataset	Method	Accuracy	Precision	Recall	F1 score
Twitter	EANN	0.623	0.547	0.747	0.631
	SpotFake	0.627	0.550	0.767	0.641
	Proposed	0.682	0.596	0.830	0.693
Weibo	EANN	0.727	0.736	0.711	0.723
	SpotFake	0.769	0.779	0.753	0.766
	Proposed	0.812	0.833	0.781	0.806