Binary Classification of Human Emotion using EEG and LTSM

Emotion detection is an important aspect in the development of AI based systems to assist humans. People affected with neurological disorders are paralyzed and have difficulty in expressing their emotions. Identifying their emotions and act accordingly helps to keep them happy and healthy. In this paper we have attempted a system that detects and classifies human emotion using Electroencephalography (EEG) signals. The process involves getting the data from the four electrodes placed on the scalp of a human and monitoring the brain activity. The algorithm uses a Long Short-Term Memory network (LTSM) which gives a training accuracy of 95.55%.