浙江大学学报（理学版）2016，Vol.43Issue(3)：357-363,7.DOI:10.3785/j.issn.1008-9497.2016.03.017
基于SVM和LS-SVM的住宅工程造价预测研究
Forecasting the costs of residential construction based on support vector machine and least squares-support vector machine
摘要
Abstract
To forecast the costs of a residential construction rapidly and accurately at the initial stage of construction that lacks relevant information ,in view of the strengths and weaknesses of previous approaches ,we choose support vector machine (SVM ) and principal component analysis (PCA) .Firstly ,a residential project cost forecasting index set is selected ;The data of the input index is then analyzed and the correlation is eliminated by PCA ;Thirdly ,the processed data are imported into the standard support vector machine and trained by the least squares support vector machine model .The prediction results are compared and analyzed ,and then a more reasonable prediction model is adopted ;Finally ,the prediction result of the model is optimized by model parameter optimization .Experiments show that the relative error of the prediction model is controlled within ± 7% ,and the result is stable .关键词
造价预测/主成分分析/支持向量机/最小二乘支持向量机Key words
construction cost forecasting/principal component analysis/support vector machine/least squares support vector machine分类
建筑与水利引用本文复制引用
秦中伏,雷小龙,翟东,金灵志..基于SVM和LS-SVM的住宅工程造价预测研究[J].浙江大学学报（理学版）,2016,43(3):357-363,7.基金项目
国网浙江省电力公司经济技术研究院资助项目（12-513205-007，名称输电线路工程造价预测快速实现）． （）
