Figure 5
From: Decoding the cognitive states of attention and distraction in a real-life setting using EEG
(a) LSTM architecture that predicted the attentional states with best accuracy. (b) Confusion matrix corresponding to test data in theta band for the dataset using LSTM which gave the maximum accuracy out of the various classifiers that were tested. (c) Comparison of classification accuracy on the basis of different data aspects for the test dataset. Delta (0.5–4 Hz) and theta (4–8 Hz) activity were most predictive of attentional state.
