Research Article

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

Table 5

Prediction performance results over the testing phase.

R2RMSEMAEMAPENashMD

MARS model
M10.7556255636.439175.30.113170.755610.77964
M20.785425216936042.40.104610.785130.79856
M30.8185347949.336928.80.106980.818480.79537
M40.8666941152.632861.10.096550.866290.82096
M50.9412527422.719761.80.054540.940630.89374
M60.9412527422.719761.80.054540.940630.89374

ELM model
M10.703896166243434.80.122040.699810.75368
M20.7268759271.241879.90.119480.722640.76358
M30.7752153666.141776.60.119910.772620.76784
M40.8004750554.140050.80.116420.798220.77885
M50.89437239.825507.30.069220.890510.86168
M60.8946337141.825376.60.068870.891090.86241

PLS model
M10.699762852.245250.30.127470.688110.74448
M20.7141160628.343182.20.123580.709790.75411
M30.7738153854.141631.90.119410.771020.76864
M40.8004750554.340051.80.116430.798220.77884
M50.894137229.5255410.069280.890570.86148
M60.8946437140.325418.70.068980.891090.86217