Deep Learning Based on CNN for Emotion Recognition Using EEG Signal

Abstract

Emotion recognition based on brain-computer interface (BCI) has attracted important research attention despite its difficulty. It plays a vital role in human cognition and helps in making the decision. Many researchers use electroencephalograms (EEG) signals to study emotion because of its easy and convenient. Deep learning has been employed for the emotion recognition system. It recognizes emotion into single or multi-models, with visual or music stimuli shown on a screen. In this article, the convolutional neural network (CNN) model is introduced to simultaneously learn the feature and recognize the emotion of positive, neutral, and negative states of pure EEG signals single model based on the SJTU emotion EEG dataset (SEED) with ResNet50 and Adam optimizer. The dataset is shuffle, divided into training and testing, and then fed to the CNN model. The negative emotion has the highest accuracy of 94.86% fellow by neutral emotion with 94.29% and positive emotion with 93.25% respectively. With average accuracy of 94.13%. The results showed excellent classification ability of the model and can improve emotion recognition.

Using Convolutional Neural Network to Recognize Emotions from Scalp Electroencephalography Plot Images

“…Upon comparing the proposed method with the existing approaches [2,7,15,16,21], it was found that our approach achieved a reduced error rate and higher accuracy. These results demonstrate the effectiveness of our approach in improving the accuracy of emotion recognition from EEG plot images.…”

“…The results of the proposed improved convolutional neural network (CNN) are presented in Table 2. A comparison was conducted between the proposed method and the methods outlined in the papers [2,7,15,16,21]. Mean accuracy of these papers is respectively 62.87, 70.50, 74.88, 82.88 and 68.11 but mean accuracy of the proposed method is 85.13.…”

“…In [21], an algorithm for EEG-based emotion detection was proposed. This algorithm decomposed emotion-related EEG signals using GaborSD functions and represented them in the spectral, spatial, and temporal domains.…”

Emotion Detection Using Deep Normalized Attention-Based Neural Network and Modified-Random Forest

“…Nevertheless, CNN has performed better than conventional techniques in categorizing EEG signals [ 20 ]. Besides, for concurrent learning of features and recognizing emotions, the study [ 21 ] has used a CNN model [ 22 ] that relies on SEED (SJTU Emotion EEG Dataset) with Adam optimizer and ResNet-50. Accuracy has been exposed to 94.13%.…”

“…Accordingly, an entropy-weighted clustering approach has been integrated with sparse learning has exposed 68.35% [ 8 ], and ANN has shown 84.3%. SVM has achieved 77.1% accuracy [ 10 ], while Rotation Forest-SVM has attained 93.1% [ 19 ], LSTM with attention Autoencoder has achieved 76.7% [ 23 ], optimized SVM reached 93.86% [ 13 ], K-NN reached 83.77%, and ANN reached 84.50% [ 25 ], and CNN has attained 94.13% [ 21 ]. Though better performance has been attained, the accuracy rate has to be enhanced for effective emotion detection [ 11 ].…”