Research Article
Developing an Integrative Data Intelligence Model for Construction Cost Estimation
Table 3
Performance measures of AI models for training data.
| ā | R2 | RMSE | MAE | MAPE | Nash | WI |
| XGBoost-RF model | Model I | 0.76081 | 788890.3191 | 604981.6715 | 0.99102 | 0.75994 | 0.93074 | Model II | 0.70272 | 879643.0411 | 589663.4386 | 0.63845 | 0.70154 | 0.90834 | Model III | 0.91381 | 472714.8598 | 339706.5592 | 0.43244 | 0.91381 | 0.97704 | Model IV | 0.92352 | 446670.1333 | 328896.3663 | 0.44469 | 0.92304 | 0.97921 | Model V | 0.96647 | 323123.8558 | 208083.017 | 0.36537 | 0.95973 | 0.98879 | Model VI | 0.93993 | 394961.8018 | 274148.5711 | 0.41215 | 0.95603 | 0.98446 | Model VII | 0.90717 | 495125.5653 | 337773.447 | 0.43415 | 0.90544 | 0.97555 |
| XGBoost-ANN model | Model I | 0.69691 | 860530.7166 | 839959.9524 | 0.77566 | 0.61666 | 0.83414 | Model II | 0.88621 | 582676.712 | 290912.5011 | 0.26372 | 0.86904 | 0.96865 | Model III | 0.95504 | 341388.8802 | 224162.0108 | 0.37779 | 0.95505 | 0.98838 | Model IV | 0.95569 | 338911.6945 | 214402.0251 | 0.40876 | 0.95569 | 0.98856 | Model V | 0.97551 | 253464.6776 | 151999.7328 | 0.40876 | 0.97522 | 0.99376 | Model VI | 0.95603 | 337627.8199 | 226863.17 | 0.41546 | 0.71921 | 0.91102 | Model VII | 0.70773 | 1108463.447 | 872191.1237 | 0.95867 | 0.52607 | 0.75929 |
| XGBoost-SVM model | Model I | 0.71722 | 838810.8081 | 666081.6149 | 0.63946 | 0.71436 | 0.91413 | Model II | 0.89679 | 527105.4427 | 333897.5918 | 0.40414 | 0.89283 | 0.97276 | Model III | 0.80479 | 717164.8809 | 466107.817 | 0.64501 | 0.80161 | 0.93934 | Model IV | 0.80151 | 735499.0773 | 491505.0568 | 0.65881 | 0.79134 | 0.93388 | Model V | 0.8031 | 715627.4657 | 451267.6691 | 0.53051 | 0.80246 | 0.94242 | Model VI | 0.72157 | 853201.115 | 542534.255 | 0.60776 | 0.79545 | 0.93821 | Model VII | 0.80124 | 727702.5318 | 450836.0521 | 0.54879 | 0.79574 | 0.93678 |
|
|