A Novel Deep Learning based Improved Cluster based Region Classifier Algorithm to Recognize and Categorize Emotions using EEG Signals

Nowadays deep learning plays vital role in emotion recognition. It distinguishes emotions as easy or multi-models for visual capturing. This works to provide an automatic version for identifying feelings primarily based on EEG signals. The proposed version specializes in developing an effective model, which combines the basic ranges of EEG signal handling and feature extraction. A system is developed based on Independent component analysis (ICA) algorithm to overcome the recognition task which removes noise object and to extract the independent components, for the obtaining components. The channels were selected based on the threshold average activity value. K-Nearest Neighbor(KNN) and Artificial Neural Network (ANN) are used to categorize emotional states and extracted the features, together with the unconventional improved Cluster-based region Classifier (ICBRC). Based on EEG signals Average recognition rate up to 94% for three emotional states and 95% for binary states can be achieved with this system.