Enhancing Emotion Detection through EEG Signal Processing with Machine Learning Techniques

Brain-Computer Interfaces (BCIs) serve as a communication bridge between an individual's brain activity and external devices. Focused on noninvasive techniques, our study delves into various methods for feature selection and extraction from EEG signals to facilitate emotion detection. The paper investigates the acquisition and preprocessing of these signals, extracting essential characteristics to reduce dimensionality. Subsequently, machine learning algorithms, including Support Vector Machine (SVM), Linear Discriminant Analysis (LDA), Fuzzy Classifiers, Regression Trees, and k-Nearest Neighbor, are examined and compared in terms of complexity and performance.