Home About Login Current Archives Announcements Editorial Board
Submit Now For Authors Call for Submissions Statistics Contact
Home > Archives > Volume 20, No 8 (2022) > Article

DOI: 10.14704/nq.2022.20.8.NQ44170

EEG Emotion Detection Using Optimum Channels

Nandini K. Bhandari, Manish Jain


Electroencephalogram (EEG) emotion recognition has been widely accepted in several applications, including intelligent thought, decision-making, behaviuor therapy, affective computing, etc. However, due to the EEG signal's insufficient amplitude change concerning time, emotion identification from this signal has become very difficult. Therefore, identifying the correct feature or feature set for a successful feature-based emotion identification system is typically substantial work. We have extracted differential entropy from five EEG channels in this proposed work. A CNN-LSTM network is used to evaluate spatio-temporal relationships from features and classify emotions. The results indicated that the 4-class classification accuracy for 32 channels is 80.2%, the accuracy for all channels situated in the frontal area is 79.52%, and the accuracy of the best four frontal channels is 76.9%. The result gives a fresh viewpoint for developing an EEG-based emotion detection system with fewer channels.


Differential Entropy, CNN (convolution neural network), EEG, emotion classification, channel selection

Full Text