东北大学学报（自然科学版）

东北大学学报(自然科学版) ›› 2011, Vol. 32 ›› Issue (5): 658-661.DOI: -

• 论著 • 上一篇    下一篇

特征选择算法在ECoG分类中的应用

刘冲;李春胜;赵海滨;王宏;   

  1. 东北大学机械工程与自动化学院;东北大学中荷生物医学与信息工程学院;
  • 收稿日期:2013-06-19 修回日期:2013-06-19 发布日期:2013-04-04
  • 通讯作者: -
  • 作者简介:-
  • 基金资助:
    国家自然科学基金资助项目(61071057)

Application of feature selection and SVM for ECoG classification

Liu, Chong (1); Li, Chun-Sheng (2); Zhao, Hai-Bin (1); Wang, Hong (1)   

  1. (1) School of Mechanical Engineering and Automation, Northeastern University, Shenyang 110819, China; (2) School of Sino-Dutch Biomedical and Information Engineering, Northeastern University, Shenyang 110819, China
  • Received:2013-06-19 Revised:2013-06-19 Published:2013-04-04
  • Contact: Liu, C.
  • About author:-
  • Supported by:
    -

RichHTML

0

PDF (PC)

772

摘要： 研究了基于运动想象的皮层脑电信号ECoG的特点,针对BCI2005竞赛数据集I中的ECoG信号,通过提取频带能量获得了想象左手小指及舌头运动时的特征,结合Fisher,SVM-RFE及L0算法对特征进行选择,采用10段交叉验证的方法得到训练数据集在各维特征数下的识别正确率并选出最佳特征组合.结果表明:三种特征选择方法中SVM-RFE算法所选出的特征组合可以获得最低的识别错误率以及最低的特征维数,针对所选出的特征组合,使用训练数据集的特征对线性支持向量机进行训练,使用训练好的模型对测试数据集进行分类,识别正确率可以达到94%.

关键词: 皮层脑电, 特征选择, 频带能量, 支持向量机, 交叉验证

Abstract: The motor imagery ECoG(electrocorticongraph) was investigated, specifically for classifying different imagined movements of the left little finger and tongue through ECoG, with BP(band power) of the ECoG signal extracted as the feature of BCI2005 competitive dataset I. Then Fisher, L0, and SVM-RFE were each used to select the best features. After a 10-fold cross validation of the training dataset, the features selected by SVM-RFE were the best of the three feature selection methods because it provided the lowest classification error rate and the least feature dimensionality. In addition, selected features of the training dataset were used to train a linear SVM model while selected features of the testing dataset were used to predict the labels by the model. Final classification accuracy was 94%.

中图分类号: