A Method to Extract Causality for Safety Events in Chemical Accidents from Fault Trees and Accident Reports

Table 2

Value of optimal parameters in the model.

Item	Value
Learning rate	0.001
Batch-size	50
Gru-size	128
Dropout	0.7
Bias constant in LN:	0.001
Number of iterations	200
Embedding size	200
Layer number	4
Regularization parameter in LN: ζ	0.0001
Bias constant in LN: t	0.001