Cost Index Predictions for Construction Engineering Based on LSTM Neural Networks

Table 7

Effects of the model’s structure on the accuracy results.

Model	Model structure	Mean absolute error	Mean square error	Accuracy (%)
M123-h10	16-10-1	1.11	1.37	99.18
M123-h20	16-20-1	1.07	1.28	99.21
M123-h30	16-30-1	1.07	1.23	99.21
M123-h40	16-40-1	0.85	0.89	99.34
M123-h50	16-50-1	0.96	1.02	99.29
M123-h60	16-60-1	1.02	1.14	99.24
M123-h70	16-70-1	1.05	1.26	99.21
M123-h80	16-80-1	1.06	1.22	99.21
M123-h90	16-90-1	1.01	1.15	99.27
M123-h100	16-100-1	1.25	1.67	99.11
M123-h110	16-110-1	1.05	1.16	99.24
M123-h120	16-120-1	1.14	1.41	99.17
M123-h130	16-130-1	1.16	1.45	99.15
M123-h140	16-140-1	1.09	1.29	99.20
M123-h150	16-150-1	1.13	1.48	99.18