In this paper,to solve the four-way classification problem of motion image EEG signals,wavelet packet decomposi-tion and common space pattern are used to extract the features,and then gradient lifting decision tree classification is used to ana-lyze the extracted features.The gradient lifting decision tree can flexibly process data of various discrete values and continuous val-ues,and does not need to normalize the data,and can directly combine features and naturally process each missing value.Com-pared with other methods,the experimental results show that the classification accuracy of this method is 0.913 7.The experimental results show that the combination of feature extraction and classification method is effective and excellent.
关键词
梯度提升决策树/运动想象脑电信号/小波分解/共空间模式
Key words
gradient lifting decision tree/motion image EEG/wavelet decomposition/common space model
维普
万方数据