Scholarlink

This research aims to develop an algorithm for accurately detecting human emotions from Electroencephalogram (EEG) signals. The study focuses on comparing the effectiveness of the Phase Slope Index (PSI), Phase Lag Value (PLV), and Empirical Mode Decomposition (EMD) domain features in classifying emotions. Additionally, a multidomain feature subset is proposed by fusing connectivity domain features with features from the Empirical Mode Domain. To improve classification performance individual feature selection method is implemented to identify the most discriminative feature subset among all extracted features. A hybrid F score-based SVM Feature Selection method has been developed for PSI and PLV-based features and an advanced correlation-based feature selection has been developed for EMD domain feature selection. Furthermore, a Stacked Ensemble Mode classifier is designed to handle features from both domains and achieve maximum accuracy in emotion classification. The outcomes of this research contribute to the development of computational models for emotion detection using EEG signals, with potential applications in affective computing.

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

Advancing EEG-Based Emotion Detection with Fusion of F-Connectivity and EMD Features and Ensemble Classification

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Sricheta Parui,
  • Srinithi Mitra,
  • Anjali Diwan

摘要

This research aims to develop an algorithm for accurately detecting human emotions from Electroencephalogram (EEG) signals. The study focuses on comparing the effectiveness of the Phase Slope Index (PSI), Phase Lag Value (PLV), and Empirical Mode Decomposition (EMD) domain features in classifying emotions. Additionally, a multidomain feature subset is proposed by fusing connectivity domain features with features from the Empirical Mode Domain. To improve classification performance individual feature selection method is implemented to identify the most discriminative feature subset among all extracted features. A hybrid F score-based SVM Feature Selection method has been developed for PSI and PLV-based features and an advanced correlation-based feature selection has been developed for EMD domain feature selection. Furthermore, a Stacked Ensemble Mode classifier is designed to handle features from both domains and achieve maximum accuracy in emotion classification. The outcomes of this research contribute to the development of computational models for emotion detection using EEG signals, with potential applications in affective computing.