Research on Attention Classification Based on Long Short-term Memory Network

Attention research is of great significance for driver attention monitoring. The biggest problem in existing attention research is that the recognition accuracy cannot be improved. The main reasons include the inability to extract the timing characteristics of the original EEG attention signal. Through designing experimental schemes, collecting original EEG data, training long short-term memory network (LSTM), and comparing various classic algorithms, the accuracy of attention recognition is further improved. Experiments show that LSTM can save the timing characteristics of the original EEG attention signal, which greatly improves the accuracy of attention recognition.