Research Article

Prediction of Concrete Compressive Strength and Slump by Machine Learning Methods

Table 5

The results of the normalization methods.

Regression methodNormalization methodD112_fc datasetD112_S datasetD224_fc datasetD224_S dataset
RMSEMAERMSEMAERMSEMAERMSEMAE

3NNMin-max5.323.9619.9318.245.513.4227.8726.34
Decimal3.132.4125.4624.793.392.7029.0927.80
Sigmoid5.213.3526.4125.215.653.6128.1326.64
Z-norm5.223.6225.8424.685.603.5028.2226.69

5NNMin-max6.535.0919.7017.995.334.6430.1728.92
Decimal2.782.3223.4422.613.463.0734.1733.12
Sigmoid6.035.2422.8221.265.504.6630.7329.56
Z-norm6.015.2222.8121.165.664.8730.3429.22

7NNMin-max5.514.3620.2519.095.514.5727.6126.53
Decimal3.602.8921.9721.263.783.1229.4028.18
Sigmoid5.694.7721.6520.705.504.2526.6325.66
Z-norm5.634.7221.6320.675.464.2026.6325.64