An Improved Long Short-Term Memory Model for Dam Displacement Prediction

Table 4

Evaluation indexes in the MLR, MLP, BRT, LSTM, and ILSTM models of L5H291R.

Model	MAE		MSE		R		Training time, s
	Training	Validation	Training	Validation	Training	Validation
MLR	1.434	1.603	3.042	3.870	0.981	0.981	0.001
MLP	1.176	1.243	2.172	2.516	0.986	0.986	0.018
SVM	1.465	1.454	3.217	3.288	0.980	0.982	0.003
BRT	0.249	1.601	0.106	3.684	0.999	0.979	1.49
LSTM	1.083	1.298	1.878	2.828	0.989	0.986	336.08
ILSTM	1.088	1.143	1.891	2.104	0.989	0.989	340.23

MAE = mean absolute error; MSE = mean square error; R = the correlation coefficient; MLR = multiple linear regression; MLP = multilayer perceptron; SVM = support vector machine; BRT = boosted regression tree. The best results are highlighted in italic.