Developing an Efficient Deep Learning-Based Trusted Model for Pervasive Computing Using an LSTM-Based Classification Model

Table 2

Training parameters of the backward propagation deep neural network.

Network	BPNN1	BPNN2	BPNN3	BPNN4	BPNN5	BPNN6
Training instances	220	220	220	220	220	220
Validating instances	100	100	100	100	100	100
Learning rate	0.001	0.0001	0.0001	0.01	0.0001	0.001
Activation function	relu	relu	relu	relu	relu	relu
Epochs	200	600	800	900	1000	1200
Training time (s)	130	150	160	165	200	250
Accuracy (%)	78.00	81.02	88.88	85.88	87.30	88.76
AUC (%)	78.23	81.45	88.78	85.78	86.22	88.76