| 引用本文: | 吴文峰,刘毅慧.基于高频系数小波分析的高维蛋白质波谱数据特征提取[J].生物信息学,2015,13(3):198-204. |
| WU Wenfeng,LIU Yihui.Feature selection forhigh-dimensional protein mass spectrometry data based on the high frequency coefficients of wavelet analysis[J].Chinese Journal of Bioinformatics,2015,13(3):198-204. |
|
| 摘要: |
| 高维蛋白质波谱数据分析过程中,对于数据的特征提取一直是许多学者专注解决的问题。本文提出了一种基于高频系数的小波分析和主成份分析技术(Principal component analysis,PCA)的特征提取方法,首先采用小波分析技术对数据进行降噪,提取高频系数作为特征,之后用主成份分析技术进行降维。实验显示:本论文中提出的方法在8-7-02、4/3/02数据集上的实验识别率分别可以达到100%和99.45%,可以有效提高分类识别率。 |
| 关键词: 波谱数据 高频 小波分析 主成份分析 |
| DOI：10.3969/j.issn.1672-5565.2015.03.09 |
| 分类号:Q629.73 |
| 基金项目: |
|
| Feature selection forhigh-dimensional protein mass spectrometry data based on the high frequency coefficients of wavelet analysis |
|
WU Wenfeng,LIU Yihui
|
|
(School of Information,Qilu University of Technology,Jinan 250353,China)
|
| Abstract: |
| During the analysis of high-dimensional protein mass spectrometry data,feature selection of the data is always the focus for many researchers.In this paper,we proposed a feature selection method based on the high frequency coefficients of wavelet analysis and principal component analysis.First we used wavelet analysis to reduct the noise,and extracted the high frequency coefficients as the feature.Then we use PCA to reduce the dimensions.The test show that when the method was applied to the data set 8-7-2,4/3/02,we can get different recognition rates of 100% and 99.45%, respectively, indicate the method can improve recognition rates effectively. |
| Key words: Spectrometry data High frequency coefficients Wavelet analysis Principal component analysis |
