Eeg-Based Mental Stress Detection Model Using Machine Learning

The aim of this research is to develop an EEG-Based model for mental stress detection using an optimal number of EEG-Channels and selecting optimal features subsets. To achieve this aim, the following sub-objective need to be accomplished: 1. To identify and extract EEG signal features and patterns that co-related to mental stress patterns using multi-domain feature set from time domain, frequency domain, time-frequency domain, and network connectivity features. 2. To develop a model for EEG stress detection with an optimal number of EEG channels using improved correlated coefficient channel selection. 3. To develop a hybrid feature selection method that can select an optimal feature subset using the hybrid swarm intelligence method.