Novel Classification System for Classifying Cognitive Workload Levels under Vague Visual Stimulation

Rwan Mahmoud, Tamer Shanableh, Indu Prasad Bodala, Nitish V. Thakor, Hasan Al-Nashash

Research output: Contribution to journalArticlepeer-review

9 Scopus citations

Abstract

This paper presents a novel method for classifying four different levels of cognitive workload. The workload levels are generated using visual stimuli degradation. Electroencephalogram (EEG) signals recorded from 16 subjects were used for workload classification. The proposed solution includes preprocessing of EEG signals and feature extraction based on statistical features. This is followed by variable selection using stepwise regression and multiclass linear classification. The presented method achieved an average classification accuracy of 93.4%. The effect of EEG channel selection on the classification accuracy is also investigated. In comparison with the existing work, we show that the proposed solution is more accurate and computationally less demanding.

Original languageEnglish (US)
Article number7982616
Pages (from-to)7019-7028
Number of pages10
JournalIEEE Sensors Journal
Volume17
Issue number21
DOIs
StatePublished - Nov 1 2017
Externally publishedYes

Keywords

  • Channel selection
  • EEG
  • cognitive workload
  • stepwise regression

ASJC Scopus subject areas

  • Instrumentation
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Novel Classification System for Classifying Cognitive Workload Levels under Vague Visual Stimulation'. Together they form a unique fingerprint.

Cite this