Zhaowen Xiao, Qingshan She, Lei Chen, Yuliang Ma
Feature selection plays a crucial role in electroencephalography (EEG)-based emotion recognition. Currently, there is limited research connecting feature selection with specific brain regions in EEG emotion recognition, and the application of transfer learning in feature selection is also scarce. This paper proposes a novel manifold sorting feature selection (MSFS) method, and its corresponding multi-source classification framework. MSFS selects channel features most similar to the T7 and T8 channels, which are highly related to each emotion category, using effective manifold-based similarity ranking. The selected features are combined and used for prediction and classification. Experimental results demonstrate that the average classification accuracy of the MSFS method combined with SVM on the SEED, and SEED-IV is 77.63%, and 53.65%, which are 1.26%, and 2.90% higher than the SVM without MSFS. Furthermore, MSFS is compared with other feature selection methods, and the results show that MSFS achieves the best performance when used in conjunction with transfer learning for classification.
Brain-computer interfaces; Feature selection; Cross-subject; Emotion recognition
DOWNLOAD FULL ARTICLE