EEG-based Emotion Recognition: An In-depth Analysis using DEAP and SEED Datasets

Research on emotion recognition has made an increasing amount of emphasis on the understanding of Electroencephalogram (EEG) signals. Using two well-known datasets - the SEED (SEED Dataset for Emotion Analysis using EEG) and the DEAP (Dataset for Emotion Analysis using Physiological Signals), this work explores the complex analysis of EEG signals and their use in emotion recognition. The study highlights important characteristics suggestive of emotional states while delving into the basic ideas behind the acquisition, processing, and interpretation of EEG signals. We explore the viability and efficiency of using EEG signals for emotion recognition tasks by utilizing machine learning and signal processing techniques. We also explore the opportunities and problems associated with EEG-based emotion identification systems, such as feature selection, artefact removal, and signal noise. The goal of this study is to offer researchers and practitioners useful insights into using EEG signals for emotion identification applications through a thorough evaluation and analysis. Showcasing the effectiveness of our methodology in EEG-based emotion recognition. The study demonstrates promising arousal accuracy at 70.88% and notable valence accuracy at 76.00% using SVM Classifier.