EMOTION RECOGNITION FROM EEG SIGNALS USING MACHINE LEARNING MODELS

Abstract

The objective of this work is to identify emotion from EEG signals, that represent the brain ac tivity of individuals. With the rapid advancement of machine learning algorithms and numerous real-world applications of brain-computer interface for regular people, emotion categorization from EEG data has recently gained a lot of attention. Researchers previously have little knowl edge of the specific interactions between distinct EEG characteristics and various emotional states. The computer may peer into the user’s head to assess the mental state with the use of EEG-based emotion identification. This work is executed with DEAP dataset with 32 channels for EEG recording, it achieves a better classification accuracy with different machine learning models. In the subject wise experiment an average best accuracies of 91.26%, 92.83% and 94.99%, and in the subject dependent experiment, the best accuracies of 78.5%, 82.77% and 92.73% is obtained for the random forest, XGBoost, and KNN classifiers respectively.