期刊文献+

投影寻踪降维方法及其试验数据分析 被引量:1

PP method and its application on test data analysis
下载PDF
导出
摘要 针对高维数据分析遇到的困难 ,引入投影寻踪降维方法进行变形观测数据分析 ,论述了分析原理、分析方法。将投影寻踪降维法应用于桥塔位移观测数据分析 ,与偏最小二乘回归法进行了分析效果对比。分析结果表明 :该方法在提取源数据信息与预测方面效果更理想。 Aimed at the difficulties coming across during the course of high dimension data analysis, a latest statistic method called as 'PP'(project pursuit) method was used. According to PP method, the high dimension data are projected to the low dimension data space, thus the projects can reflect the feature or structure of original high dimension data. By investigating the project in low dimension data space, the effective multi-factors regression analysis on deformation is made, 'PP' method is applied to test data regression analysis of new bridge before acceptance. Compared with the results derived from partial least square regression and neural network analysis on test data analysis on new bridge, the 'PP' method is proved to be a practical way in some aspects of obtaining original data information and forecasting.
出处 《中国有色金属学报》 EI CAS CSCD 北大核心 2003年第3期743-748,共6页 The Chinese Journal of Nonferrous Metals
基金 霍英东教育基金会青年教师基金资助项目 (710 17) 国家自然科学基金资助项目 (40 0 0 10 17) 湖南省自然科学基金资助项目(0 2JJY2 0 6 4)
关键词 高维空间 投影寻踪 变形观测 数据分析 桥塔 偏最小二乘回归法 multi-dimension space project pursuit deformation observation data analysis
  • 相关文献

参考文献3

  • 1[1]HUANG Quan-yi, ZHANG Zheng-lu. Dam deformation prediction by artificial neural network method[A]. Proceedings of Monitoring of Constructions and Local Geodynamic Process[C].Wuhan: Wuhan University Press, 2001. 123-128.
  • 2[2]Kropp J. A neural network approach to the analysis of city system[J]. Applied Geography, 1998, 18(1): 83-96.
  • 3[7]Friedman J H, Stuetzle W. Projection pursuit regression[J]. Journal of American statistical Association, 1981(76): 817-823.

同被引文献5

引证文献1

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部