Scholarlink

<p>Electroencephalography (EEG) has proven to be a reliable technique for recognizing human emotions, with applications in defense, aerospace, and mental health. This study presents an automated EEG-based emotion recognition framework that classifies emotional states using seven classical machine learning algorithms: Decision Tree (DT), Random Forest (RF), AdaBoost, Gaussian Naive Bayes (GNB), Linear Discriminant Analysis (LDA), Logistic Regression (LR), and K-Nearest Neighbors (KNN). Experiments conducted on the GAMEEMO, DEAP, and DREAMER datasets reveal that ensemble and probabilistic models deliver superior results. RF achieved the highest accuracy of 99.99% on GAMEEMO and 99.96% on DEAP, while GNB attained 99.36% accuracy on Dreamer with minimal computation time. DT also demonstrated strong accuracy with excellent efficiency, making it suitable for real-time applications. Overall, the proposed framework achieves up to 99.99% accuracy, offering a robust benchmark for EEG-based emotion classification. These results highlight the effectiveness of ensemble and probabilistic approaches in capturing nonlinear neural patterns, advancing the precision and practicality of affective computing systems.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

EEG Signal-Based Emotion Recognition Using Machine Learning Algorithm

  • 期刊论文

  • 发表日期：

    2026年3月30日

查看全文
  • Akhilesh Kumar,
  • Awadhesh Kumar

摘要

Electroencephalography (EEG) has proven to be a reliable technique for recognizing human emotions, with applications in defense, aerospace, and mental health. This study presents an automated EEG-based emotion recognition framework that classifies emotional states using seven classical machine learning algorithms: Decision Tree (DT), Random Forest (RF), AdaBoost, Gaussian Naive Bayes (GNB), Linear Discriminant Analysis (LDA), Logistic Regression (LR), and K-Nearest Neighbors (KNN). Experiments conducted on the GAMEEMO, DEAP, and DREAMER datasets reveal that ensemble and probabilistic models deliver superior results. RF achieved the highest accuracy of 99.99% on GAMEEMO and 99.96% on DEAP, while GNB attained 99.36% accuracy on Dreamer with minimal computation time. DT also demonstrated strong accuracy with excellent efficiency, making it suitable for real-time applications. Overall, the proposed framework achieves up to 99.99% accuracy, offering a robust benchmark for EEG-based emotion classification. These results highlight the effectiveness of ensemble and probabilistic approaches in capturing nonlinear neural patterns, advancing the precision and practicality of affective computing systems.