Comprehensive Analysis of Feature Extraction and Classification Methods for Emotion Recognition using EEG Signals

Abstract

The decision capacity of human brain based on emotion. Emotion changes the physical behaviours of human body and reaction activity. The acceleration of sensors enable technology developed brain computer interface and applied for emotion recognition. Now a days various nervous related disease diagnoses with the help of emotion recognition. In this paper study and analysis of emotion recognition based on EEG signals. The nature of EEG signals is very diverse and high dimension. For the processing of EEG data applied various feature extraction methods. the feature extraction methods extract the different bands of features for the processing of classification and recognition. The process of recognition applied various classification algorithm such as SVM, KNN, DT, NB and CNN. For the validation of classification algorithms use DEPA dataset. The DEPA dataset is free available for the study purpose. For the evaluation of the performance of emotion recognition measure standard parameters such as precision, recall and F-measure. For the process of simulation use MATLAB software.