| 基于多元图形特征融合原理的降维方法研究 |
| |
| 引用本文： | 孟辉,洪文学,宋佳霖,王立强. 基于多元图形特征融合原理的降维方法研究[J]. 燕山大学学报, 2008, 32(5) |
| |
| 作者姓名： | 孟辉 洪文学 宋佳霖 王立强 |
| |
| 作者单位： | 1. 燕山大学,电气工程学院,河北,秦皇岛,066004
2. 燕山大学,车辆与能源学院,河北,秦皇岛,066004 |
| |
| 基金项目： | 国家自然科学基金，教育部高等学校博士学科点专项科研基金，河北省自然科学基金 |
| |
| 摘 要： | 降维是将高维模式映射到低维子空间的过程.在降维后的低维子空间进行分类往往能得到更好的效果.本文以高维数据为研究对象,采用多元描述图对高维数据进行可视化表达,采用多元图图形特征融合的方法对高维数据进行降维,用K邻分类器进行分类效果评价.与Fisher线性判别及其他一些常用非线性降维方法相比,本文所提方法在数据的可视化以及分类精度等方面均有较好效果.
|
| 关 键 词： | 降维 多元图形特征融合 雷达图 |
A new method for dimensionality reduction based on multivariate graphical feature fusion |
| |
| MENG Hui,HONG Wen-xue,SONG Jia-lin,WANG Li-qiang. A new method for dimensionality reduction based on multivariate graphical feature fusion[J]. Journal of Yanshan University, 2008, 32(5) |
| |
| Authors: | MENG Hui HONG Wen-xue SONG Jia-lin WANG Li-qiang |
| |
| Affiliation: | MENG Hui 1,HONG Wen-xue 1,SONG Jia-lin 1,WANG Li-qiang 2 (1. College of ElectricalEngineering,Yanshan University,Qinhuangdao,Hebei 066004,China,2. College of Vehicle ,Energy Engineering,China) |
| |
| Abstract: | Dimensionality reduction is the process of mapping high-dimension patterns to a lower dimension subspace. When done prior to classification, estimates obtained in the lower dimension subspace are more reliable. A novel method based on graphical multivariate feature fusion is proposed and used to offer a visual representation of high dimensional data. The graphical processing method proposed in this paper, relies on using a multilayered structure of feature fusion which produces as output of the lower di- me... |
| |
| Keywords: | dimensionality reduction multivariate graphical feature fusion star plot |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
