首页 | 本学科首页   官方微博 | 高级检索  
   检索      

基于小波低频系数基因芯片数据的特征提取
引用本文:刘玉杰,刘毅慧.基于小波低频系数基因芯片数据的特征提取[J].生物信息学,2011,9(3):255-258,262.
作者姓名:刘玉杰  刘毅慧
作者单位:山东轻工业学院信息科学与技术学院智能信息处理研究所,山东,济南,250353
摘    要:特征提取和分类是模式识别中的关键问题。结合小波分析理论和支持向量机理论,构造分类器模型,将前列腺癌基因芯片数据分成癌症和正常两种。提取小波低频系数表征原始数据并送入支持向量机分类器分类,实验证明:提取db1小波4层分解下的低频系数,送入分类器分类后正确分类率达到93.53%。Haar小波的正确率是92.94%。可见提取不同小波低频系数,得到的分类效果相差不大。

关 键 词:小波分析  支持向量机  前列腺癌基因芯片数据  交叉验证  小波低频系数

Wavelet low-frequency Coefficients for Feature Extraction of Gene Microarray data
LIU Yu-jie,LIU Yi-hui.Wavelet low-frequency Coefficients for Feature Extraction of Gene Microarray data[J].China Journal of Bioinformation,2011,9(3):255-258,262.
Authors:LIU Yu-jie  LIU Yi-hui
Institution:LIU Yu-jie,LIU Yi-hui*(Institute of Intelligence Information Processing,School of Information Science and Technology,Shandong Institute of Light Industry,Jinan 250353,China)
Abstract:Feature extraction and classification are the key issues in the pattern recognition field.In the paper,we use the wavelet analysis theory and the support vector machine theory to build a classifier which can distinguish cancer tissue from prostate cancer gene microarray data.We extract wavelet low-frequency coefficients to characterize the features of prostate cancer gene microarray data,then feed the coefficients to the classifier.In these experiments,we extract db1 wavelet low-frequency coefficients at le...
Keywords:wavelet analysis  support vector machine  prostate cancer gene microarray data  cross-validation  low-frequency coefficients  
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号