Table 2.

significant DL models comparison on standard dataset

Year	DL models applied	Datasets	Performance
2016	RNN [61]	Twitter, Sina Weibo	Accuracy$-$91.0% (Twitter) Accuracy$-$88.1% (Weibo)
2018	RNN [15]	Twitter, Weibo	Precision$-$74.02% (Twitter), 71.7% (Weibo), Recall$-$68.75% (Twitter), 70.34% (Weibo)
2019	BiLSTM, CNN[6]	PHEME	Accuracy$-$86.1%
2019	LSTM, CNN[57]	Sina Weibo	Accuracy–94% precision-93% recall-95% F1-94%
2019	GCN [33]	Twitter 15, Twitter 16	Accuracy77.3% (Twitter 15) Accuracy$-$75.2% (Twitter 16)
2019	RNN [54]	RumorEval, PHEME	Accuracy$-$63.8% (RumorEval) Accuracy$-$48.3% (PHEME)
2020	BiLSTM [86]	PHEME	Accuracy$-$92.6%(PHEME 2017) Accuracy$-$91.9% (PHEME2018)
2020	GCN [11]	Sina Weibo, Twitter 15, Twitter 16	Accuracy$-$96.1% (Sina Weibo) Accuracy$-$88.6% (Twitter 15) Accuracy$-$88.0% (Twitter 16)
2020	CNN [27]	YELP-2, FBN	Accuracy$-$87.2% precision$-$79.1% recall$-$84.7% F1-82%
2020	Deep learning [48]	PHEME	Accuracy$-$94.9% precision$-$37.4% recall$-$51.8% F1-79%
2020	Deep learning, CNN [5]	PHEME	Accuracy–94%
2021	Deep learning, LSTM [94]	PHEME	Accuracy–81% precision-79% recall-79% F1-81%
2021	GCN [93]	Twitter 15, Twitter 16	Accuracy$-$87.8% (Twitter 16) Accuracy$-$85.6% (Twitter 15)
2021	Propagation structure, CNN [90]	Twitter 15, Twitter 16, Weibo	Accuracy$-$95.1% precision$-$94.5% recall$-$95.6% F1$-$95.0%
2021	GCN, [97]	PHEME, Twitter 15, Twitter 16	Accuracy–89% (Twitter 15) Accuracy$-$91.5% (Twitter 16) Accuracy-69% (PHEME)
2021	GCN [9]	PHEME	Accuracy$-$84.1% Precision$-$88.2% Recall$-$95.6% F1- 89.6%
2021	Naive Bayes classifier [50]	PHEME	Accuracy$-$76.7%, precision$-$76.1% recall$-$76.3%
2021	Deep neural networks [8]	DAT@Z20, Fake News Net	Accuracy-94% precision-93% recall-95%
2021	CNN, LSTM [4]	ArCOV-19	Accuracy–85% precision-85% recall-85% F1-85%
2022	GCN [106]	Twitter 15, Twitter 16	Accuracy$-$86.5% (Twitter 16) Accuracy$-$83.6% (Twitter 15)
2022	BiLSTM [14]	Weibo	Accuracy-95% precision$-$94.3% recall$-$94.1%