Feature Mapping using PCA, Locally Linear Embedding and Isometric Feature Mapping for EEG-based Brain Computer Interface

To enable the user to visualize their own brain activity is an essential part of a Brain-Computer Interface (BCI). For visualization, we need a dimension reduction algorithm, because only a few relevant components of the high-dimensional feature vectors, extracted from the electroencephalogram (EEG) signals, can be visualized. For this purpose, three feature mapping methods, Principal Component Analysis (PCA), Locally Linear Embedding (LLE) and Isometric Feature Mapping (Isomap) were investigated and compared to each other.