EEG Based Human Emotion Analysis

Emotions accompany everyone in their daily lives, play an important role in nonverbal communication, and are critical to understanding human behaviours. Text, speech, facial expressions, and gestures can all be used to recognise emotions. In this paper we propose a model using electroencephalograph signal that are decomposed into different frequency bands and spectral features are extracted from them. In the proposed solution the dataset has been trained and tested using different machine learning algorithm to get all the possibilities and greater accuracy. The dataset testing mainly involved three emotions and has been tested using random forest algorithm. Using the dataset samples are collected and for each sample a decision tree is built to produce the result. In this paper various frequency bands are used with different naming with a primary objective to verify whether the naming makes a specific difference.