Forecasting the costs of residential construction based on KPCA-WOA-SVM
Accurately predicting the cost of residential engineering during the project decision-making stage is crucial for improving the scientificity of project decision-making.The development of artificial intelligence and machine learning can further improve the accuracy of predictions.Through literature research,the impact indicators of residential engineering cost in the decision-making stage are sorted out.kernel principal component analysis(KPCA)is used to reduce the dimensionality of the impact indicators,and whale optimization algorithm(WOA)is used to determine the penalty and kernel parameters of support vector machine(SVM).Finally,a residential engineering cost prediction model based on KPCA-WOA-SVM is constructed.The model validation is conducted on 70 sets of residential engineering cost data in Jiangsu Province over the past five years.The results showe that compared with the BP neural network model,SVM model,and WOA-SVM model,the KPCA-WOA-SVM model proposed in this paper has higher prediction accuracy and better applicability.
residential engineering costkernel principal component analysiswhale optimization algorithmsupport vector machine
万方数据
维普
