The Reseaches on Mixtures of Kernels Function Initial Parameters Selection on the Synthetic Evaluation Function of KPCA

摘要：

This paper study on the mixtures of kernels function initial parameters selection in the synthetic evaluation function of kernel principal component analysis (KPCA) based on the samples and kernel principal component contribution rate adaptive. It provides an effective method to overcome the shortcomings of kernels parameters difficult to selection when KPCA is applied to synthetic evaluation based on multiple indicators.

关键词： mixtures of kernels function KPCA synthetic evaluation

作者: TAN Yili ZHAO Ye WAN Xinghuo

作者单位: School of Science, Hebei Polytechnic University, Tangshan, P.R.China, 063009 School of Science, Shijiazhuang Tiedao University, Shijiazhuang, P.R.China, 050043

会议类型: 国际会议

会议名称: The 3rd International Institute of Statistics & Management Engineering Symposium(2010 国际统计与管理工程研讨会 IISMES)

会议地点: 威海

会议语种:英文

页码: 641-644

在线出版日期: 2010-07-24（万方平台首次上网日期，不代表论文的发表时间）