Research Article

Evaluation of Several Machine Learning Models for Field Canal Improvement Project Cost Prediction

Table 4

Prediction performance results over the training phase.

R2RMSEMAEMAPENashMD

MARS model
M10.7657097.2140090.040.110.760.78
M20.7952443.1135922.940.100.790.80
M30.8346957.4535377.610.100.830.81
M40.8741427.3331383.750.090.870.83
M50.9428458.1720444.550.060.940.89
M60.9428458.1720444.550.060.940.89

ELM model
M10.7261144.0042644.280.120.720.76
M20.7558309.9140861.500.120.750.77
M30.7952776.6540372.480.110.790.78
M40.8150951.9839296.320.110.810.79
M50.9036019.9424352.690.070.900.87
M60.9036011.4324362.290.070.900.87

PLS model
M10.7261182.3143650.760.120.720.75
M20.7359634.1042063.220.120.730.76
M30.7952816.7340059.100.110.790.78
M40.8150951.9839297.600.110.810.79
M50.9036026.9724406.130.070.900.87
M60.9036013.1624393.670.070.900.87