Classification of user competency levels using EEG and convolutional neural network in 3D modelling application

Muhammad Zeeshan Baig*, Manolya Kavakli

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

3 Citations (Scopus)

Abstract

Competency classification is one of the main challenging tasks for the development of state-of-the-art next generation computer-aided design (CAD) system. To develop a futuristic system that can accommodate the lack of competency, the system needs to adapt to the competency level of the user. To solve this problem, we have presented a deep convolutional neural network (CNN) model that uses the Electroencephalography (EEG) of the user to classify the level of competency in 3D modeling task. The five competency levels were defined based on the task completion time, final 3D model rating and previous modeling experience. This is the first study that classifies user competency and employs the CNN model for the analysis of EEG signals in the design application. In this work, a 14-layer deep CNN model was implemented to classify competency into five different levels. The proposed technique achieved an accuracy, specificity, and sensitivity of > 88%, > 90% and > 70% respectively with 5-fold cross-validation. The results showed the applicability of a CNN model to classify the user competency and can be used as a first step in developing state-of-the-art adaptive 3D modeling systems.

Original languageEnglish
Article number113202
Pages (from-to)1-10
Number of pages10
JournalExpert Systems With Applications
Volume146
DOIs
Publication statusPublished - 15 May 2020

Keywords

  • Deep neural networks
  • Competency
  • EEG
  • CNN
  • Novice
  • Entropy

Fingerprint

Dive into the research topics of 'Classification of user competency levels using EEG and convolutional neural network in 3D modelling application'. Together they form a unique fingerprint.

Cite this