A Novel Stress State Assessment Method for College Students Based on EEG

Area of Science:

• Neuroscience
• Machine Learning
• Psychology

Background:

• College students experience unique stressors including academic, social, and employment pressures.
• Limited coping skills and social support exacerbate mental stress in this demographic.
• Accurate stress assessment is crucial for preventing psychological issues among students.

Purpose of the Study:

• To develop and validate an effective model for assessing college student stress levels.
• To address the challenge of imbalanced datasets common in stress detection.
• To improve the performance of stress recognition models using advanced machine learning techniques.

Main Methods:

• An improved extreme learning machine (IELM) algorithm was developed, incorporating label weighting and AdaBoost.
• The IELM algorithm was applied to classify electroencephalogram (EEG) data for stress level determination.
• The method leveraged the imbalanced nature of multi-label datasets during weight updates.

Main Results:

• The IELM algorithm demonstrated excellent classification performance in assessing student stress.
• The model accurately identified stress levels in college students.
• Experimental results confirmed the efficacy of the proposed approach for stress detection.

Conclusions:

• The developed IELM model provides a reliable method for accurately assessing college student stress.
• Effective stress assessment is vital for promoting student mental health and well-being.
• This research offers significant practical implications for mental health interventions in higher education.