Research on Construction cost Prediction of Construction Project Based on Extreme Iearning Machine
With the development of the construction engineering industry,construction cost prediction has been paid more and more attention.To improve the prediction accuracy,we propose a fusion prediction model based on extreme learning machine(ELM)and support vector machine(SVM).First,we construct a sub-prediction model based on ELM and SVM respectively,and then combine the two seed models by a weighted fusion strategy.The experimental results show that the proposed fusion prediction model is superior in the construction cost prediction,and has high prediction accuracy and stability.We compare the performance of ELM,SVM and multi-layer perceptron(MLP),and analyze the fusion prediction model in detail.In addition,we also discussed the future research direction,with a view to promoting the development of the construction cost forecasting field of construction engineering.
construction engineeringconstruction cost forecastextreme learning machinesupport vector machinefusion prediction model
NETL
NSTL
万方数据
