Novel Approach for Emotion Detection and Stabilizing Mental State by Using Machine Learning Techniques

The aim of this research study is to detect emotional state by processing electroencephalography (EEG) signals and test effect meditation music therapy stabilize mental state. This useful identify 12 subtle emotions angry (annoying, angry, nervous), calm (calm, peaceful, relaxed), happy (excited, happy, pleased), sad (sleepy, bored, sad). A total 120 emotion were collected using Emotive 14 channel EEG headset. Emotions are elicited three types stimulus thoughts, audio video. system trained captured database which include 30 each class. 24 features extracted performing Chirplet transform. Band power ranked as the prominent feature. multimodel approach classifier used classify emotions. Classification accuracy tested for K-nearest neighbor (KNN), convolutional neural network (CNN), recurrent (RNN) deep (DNN) classifiers. intellectually disable people. Meditation stable It found that it changed both disabled normal participants from annoying relaxed 75% positive transformation obtained in therapy. presents a novel detailed analysis brain detection