本文選題：腦電信號 + 高斯過程分類��； 參考：《傳感技術(shù)學(xué)報(bào)》2017年03期

【摘要】：針對傳統(tǒng)的高斯過程采用共軛梯度法確定超參數(shù)時(shí)對初值有較強(qiáng)依賴性且易陷入局部最優(yōu)的問題,提出了一種基于人工蜂群優(yōu)化的高斯過程分類方法,用于腦電信號的模式識別。首先,構(gòu)建高斯過程模型,選擇合適的核函數(shù)且確定待優(yōu)化的參數(shù)。然后,選取識別錯(cuò)誤率的倒數(shù)為適應(yīng)度函數(shù),使用人工蜂群算法搜索尋找出限定范圍內(nèi)可以取得最優(yōu)準(zhǔn)確率的超參數(shù)。最后,采用參數(shù)優(yōu)化后的高斯過程分類器對樣本分類。分別采用2008年競賽數(shù)據(jù)集BCI CompetitionⅣData Set 1和2005年數(shù)據(jù)集BCI CompetitionⅢData SetⅣa對所提方法進(jìn)行驗(yàn)證,并與支持向量機(jī)(SVM)、人工蜂群優(yōu)化的支持向量機(jī)(ABC-SVM)、高斯過程分類(GPC)方法進(jìn)行比較,實(shí)驗(yàn)結(jié)果表明了所提方法的有效性。
[Abstract]:In order to solve the problem that the traditional Gao Si process has strong dependence on the initial value and is easy to fall into local optimum when using conjugate gradient method to determine the superparameters, a Gao Si process classification method based on artificial bee colony optimization is proposed. Pattern recognition for EEG signals. Firstly, the Gao Si process model is constructed, the appropriate kernel function is selected and the parameters to be optimized are determined. Then, the inverse of the recognition error rate is selected as the fitness function, and the artificial bee colony algorithm is used to search for the super-parameters which can obtain the best accuracy in the limited range. Finally, the Gao Si process classifier with optimized parameters is used to classify the samples. The proposed method is validated by BCI Competition 鈪，

本文編號：2012688