摘要
基于数据挖掘法对设备安装工程造价影响因素进行深入分析,确定了以深度神经网络(DNNs)为核心的机器学习算法,建立了设备安装工程造价预测指标(输入指标)与设备安装造价(输出指标)之间的相互关系。采用最小二乘支持向量机(LSSVM)对DNNs模型及SVM模型预测结果的风险偏差进行了评估。结果表明,采用的DNNs模型的预测结果及风险偏差分析结果均优于传统的SVM模型,DNNs模型能够对建筑工程造价进行有效预测。
Based on the data mining method,the influencing factors of equipment installation project cost are deeply analyzed,and the machine learning algorithm with deep neural network(DNNs)as the core is determined,and the relationship between equipment installation project cost prediction index(input index)and equipment installation cost(output index)is established.Least Squares Support Vector Machine(LSSVM)was used to evaluate the risk bias of the prediction results of DNNs model and SVM model.The results show that the prediction results and risk deviation analysis results of the DNNs model are better than those of the traditional SVM model,and the DNNs model can effectively predict the cost of construction projects.
作者
李丹平
张秀秀
王显亮
Li Danping;Zhang Xiuxiu;Wang Xianliang
出处
《江苏建材》
2025年第1期156-158,共3页
Jiangsu Building Materials
关键词
数据挖掘
设备安装
造价预测
偏差分析
data mining
equipment installation
cost forecasting
deviation analysis
