Dynamically weighted ensemble classification for non-stationary EEG processing

Sidath Ravindra Liyanage, Cuntai Guan, Haihong Zhang, Kai Keng Ang, JianXin Xu, Tong Heng Lee
2013 Journal of Neural Engineering  
The non-stationary nature of electroencephalogram (EEG) poses a major challenge for the operation of a brain-computer interface (BCI). This paper proposes a novel Dynamically Weighted Ensemble Classification (DWEC) framework to address the non-stationarity. An ensemble of multiple classifiers are trained on clustered features. The decisions from these multiple classifiers are dynamically combined based on the distances of the cluster centres to each test data sample being classified. The
more » ... -to-session performance of the proposed DWEC method was evaluated on two datasets. The results on publicly available BCI Competition IV dataset 2A yielded significantly higher mean accuracy of 81.48% compared to 75.9% from the baseline classifier. Results on data collected from our own 12 subjects yielded significantly higher mean accuracy of 73% compared to 69.4% from the baseline case. Hence, the results demonstrated the ability of the proposed method to address non-stationarity in EEG data for the operation of a BCI.
doi:10.1088/1741-2560/10/3/036007 pmid:23574821 fatcat:mrygve3tcjdf3n5qf3i436zvea