Classifying EEG for Brain Computer Interfaces Using Gaussian Process

Mingjun Zhong (1) , Fabien Lotte (1) , Mark Girolami (2) , Anatole Lécuyer (1)

1 BUNRAKU - Perception, decision and action of real and virtual humans in virtual environments and impact on real environments
2 School of Computing Science [Glasgow]

Classifying electroencephalography (EEG) signals is an important step for proceeding EEG-based brain computer interfaces (BCI). Currently, kernel based methods such as sup- port vector machine (SVM) are the state-of-the-art methods for this problem. In this paper, we apply Gaussian process (GP) classfi¯cation to binary classi¯cation problems of motor imaging EEG data. Comparing with SVM, GP based method naturally provides a predic- tive probability for identifying a trusted prediction which can be used for post-processing in a BCI. Experimental results show that the classification methods based on Gaussian process outperform SVM and K-Nearest Neighbor (KNN) in terms of 0-1 loss class prediction error.

Mots clés: brain-computer interface; EEG; classification; motor imagery; gaussian process; support vector machine

Domaines: Intelligence artificielle [cs.AI]; Biotechnologie; Biotechnologies; Apprentissage [cs.LG]

Mingjun Zhong, Fabien Lotte, Mark Girolami, Anatole Lécuyer. Classifying EEG for Brain Computer Interfaces Using Gaussian Process. Pattern Recognition Letters, 2007. ⟨inria-00134966v2⟩