An Investigation of Deep Learning Models for EEG-Based Emotion Recognition

Abstract

Emotion is the human brain reacting to objective things. In real life, human emotions are complex and changeable, so research into emotion recognition is of great significance in real life applications. Recently, many deep learning and machine learning methods have been widely applied in emotion recognition based on EEG signals. However, the traditional machine learning method has a major disadvantage in that the feature extraction process is usually cumbersome, which relies heavily on human experts. Then, end-to-end deep learning methods emerged as an effective method to address this disadvantage with the help of raw signal features and time-frequency spectrums. Here, we investigated the application of several deep learning models to the research field of EEG-based emotion recognition, including deep neural networks (DNN), convolutional neural networks (CNN), long short-term memory (LSTM), and a hybrid model of CNN and LSTM (CNN-LSTM). The experiments were carried on the well-known DEAP dataset. Experimental results show that the CNN and CNN-LSTM models had high classification performance in EEG-based emotion recognition, and their accurate extraction rate of RAW data reached 90.12 and 94.17%, respectively. The performance of the DNN model was not as accurate as other models, but the training speed was fast. The LSTM model was not as stable as the CNN and CNN-LSTM models. Moreover, with the same number of parameters, the training speed of the LSTM was much slower and it was difficult to achieve convergence. Additional parameter comparison experiments with other models, including epoch, learning rate, and dropout probability, were also conducted in the paper. Comparison results prove that the DNN model converged to optimal with fewer epochs and a higher learning rate. In contrast, the CNN model needed more epochs to learn. As for dropout probability, reducing the parameters by ~50% each time was appropriate.

Keywords: CNN (convolutional neural network); CNN-LSTM; DNN (deep neural network); EEG; emotion recognition.

Figure 1

International 10–20 standard system electrode position distribution map (the dark-colored ones are the channels used in this experiment).

Figure 2

Valence/arousal measurement and one-hot encoding labels.

Figure 3

The convolution-max process diagram.

Figure 4

The convolution-max process diagram.

Figure 5

The CNN model structure.

Figure 6

The schematic diagram of the LSTM unit.

Figure 7

The connection mode and working principle of the LSTM model.

Figure 8

The connection mode and working principle of the CNN-LSTM model.

Figure 9

Performance comparison histogram between different models. (A) Classfication accuracy of different models. (B) Classification loss of different models.