A CNN-LSTM Framework for EEG-Based Emotion Recognition

The advancement of deep learning architectures has facilitated the effective development of Emotion recognition, algorithms for EEG based emotion recognition. In this paper, we present a CNN-LSTM model, in which a ۱D-CNN is employed to reduce dimensionality and extract meaningful features before passing them to the LSTM, leading to improvements in both computational efficiency and recognition accuracy. The selection of an appropriate window size for feature extraction, combined with the hybrid model architecture, provides superior recognition performance compared to using CNN or LSTM alone. Experimental evaluations on the DEAP dataset confirm the effectiveness of the proposed approach.

کلیدواژه ها:

EEG ، CNN-LSTM ، deep learning ، Emotion recognition