EEG-EmoNet: A Deep Learning Framework for Emotion Recognition from EEG Data

Emotion recognition is a basic aspect of human interaction and understanding In recent years, there has been a growing interest in developing automated systems capable of accurately detecting and classifying human emotions using electroencephalography (EEG) data. EEG, a non-invasive technique for measuring electrical activity in the brain, offers valuable insights into the neural correlates of emotional states. This paper aims to explore and develop a robust emotion recognition system utilizing EEG data. An EEG dataset comprising recordings from individuals experiencing different emotional stimuli is collected, ensuring diversity and representativeness of emotions. The collected data is then pre-processed to remove noise, artifacts, and other unwanted signals, enhancing the quality and reliability of the EEG signals. The paper also explores the potential for real-time emotion recognition from EEG data, aiming to develop an efficient system that can provide timely and accurate emotion detection. We start by providing an overview of EEG and its use in emotion detection. We then review various machine learning algorithms that have been used for emotion detection using EEG data, including Gated Recurrent Unit (GRU), Attention Model etc, thus giving a test accuracy of 97.50 percent