Emotion Classification Based On EEG Signals In a Stable Environment

Emotion defines as human behavior used to know the feeling, expression sentiment of humans inside the brain. So Emotion can explain its statics through voice interaction, words, facial expression, and body language. The main objective of this paper is to analyze the accuracy of the responded data by electroencephalograph (EEG) through machine learning algorithms. After predicting the accuracy we are trying to classify results with classification learning techniques. So here we have considered different lobes points of a human brain by using the NMX32 device for converting EEG signal to raw data format. From different types of lobes point we are using two pairs of points(P3-O1, T4-T6). We used Naive Bayes and Decision trees to classify and accuracy of our data in a stable environment with different situations. The above techniques give better accuracy on the EEG signal data source as 95.4% with ROC, 83.2% with Precision, 83.1% with Recall.