Journal of Cloud Computing

Advances, Systems and Applications

Table 5 Result of various improved recurrent neural networks with the same parameters

From: Low-cost and high-performance abnormal trajectory detection based on the GRU model with deep spatiotemporal sequence analysis in cloud computing

Model	Precision	Accuracy	Recall	F1 Score	Cost(s)
Improved RNN	0.988095	0.905263	0.95203	0.928058	40
Improved GRU	0.994643	1	0.936842	0.967391	69
Improved LSTM	0.978869	0.8	0.942149	0.865275	63
Improved RNN-GRU	0.986012	0.898246	0.934307	0.915921	51
Improved RNN-LSTM	0.971131	0.870175	0.805195	0.836425	49
Improved GRU-LSTM	0.974702	0.891228	0.824675	0.856661	67

Back to article page