PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
February 19, 2026PeerJ Computer Science0 citationsOpen Access

Enhanced Swarm Optimization for Feature Selection in EEG Classification

Enhanced swarm optimization for feature selection in electroencephalogram classification: investigating visibility graph and persistent homology-based features

View Full Paper
Ask AI
Bookmark
Share

Authors

CLCarey Yu-Fan LingPPPiau PhangSLSiaw-Hong Liew

Discussion

Loading...

Member takes

Overview

Novel pipeline improves EEG classification accuracy in visual tasks, suggesting advancements in brain-computer interfaces.

Key Points

  • This research aims to enhance feature selection methods for classifying non-medical EEG data using persistent homology and swarm optimization techniques.
  • Developed a persistent homology pipeline integrating visibility graphs and enhanced binary particle swarm optimization.
  • Tested on non-medical EEG recordings under varying auditory conditions.
  • Analyzed various PH representations and filtrations for feature extraction.
  • Achieved a 23.71% increase in accuracy and a 17.77% increase in F1-score when classifying alpha EEG.
  • Enhanced BPSO outperformed standard BPSO in classification tasks.
  • Specific PH features consistently outperformed other methods in performance.

Cite This Study

Ling et al. (2026) studied this question.

synapsesocial.com/papers/6996a7b5ecb39a600b3edb19https://doi.org/10.7717/peerj-cs.3617
View Full Paper
Ask AI
Bookmark
Share