基于智能化信息处理的建筑工程造价短期预测

提出了一种新的基于智能化信息处理的建筑工程造价短期预测模型.该模型首次利用数据的高阶统计信息,提出了改进独立分量分析技术.通过构建适用于建筑工程造价的属性重构空间,挖掘出表征能力更强的造价独立属性,用于神经网络的学习和训练,从而建立了全新的建筑工程造价短期预测智能模型.该模型通过发挥独立分量分析强大的信号分离能力,增强了神经网络的学习效率,提高了预测精度.实例数据验证了文中所建模型的有效性.

A Novel Short-Term Forecasting Model of Construction Cost Based on Intelligent Information Processing Technology

A novel short-term forecasting model of construction cost based on intelligent information processing(IIP) technology is presented.By utilizing higher-ranked statistical information,independent component analysis(ICA) is improved and used to build feature reconstruction space.Then cost independent features are reconstructed to eliminate the interior correlation and redundancy.After artificial neural network is trained by these obtained independent features with improved representation ability,forecasting model of construction cost is built.The proposed model not only makes a good use of the signal separation ability of ICA,but also raises the ANN efficiency,which enhances the forecasting accuracy.Finally,the real construction cost and its relative data are employed to testify the validity of the proposed approach.