Implementation of Cluster analysis for Learning Style classification using brain Asymmetry

Abstract

This study highlighted the use of Cluster analysis approach to classify the participants' Learning Style (LS) based on the EEG brain asymmetry (BA) dataset. BA is importance to indicate brain activity in both right hemisphere (RH) and left hemisphere (LH). The RH and LH dominant states are closely related to human learning traits such as Attention, Perception and Emotions. In this research, we determine the LS of 41 participants using Kolb's Learning Style Inventory (LSI). The LSI will group them into the LS of either Diverger, Assimilator, Converger or Accommodator. Simultaneously, their Electroencephalogram (EEG) is recorded from which the BA will be calculated using the Asymmetry Relation Ratio (ARR) formula. The Alpha and Beta Energy Spectral Density (ESD) are used as input for the ARR. Finally, the SPSS 2Steps cluster analysis will be deployed to classify the BA towards the corresponding LS. The result obtained shown that the classification of each LS is achieving 100% accuracy. We also managed to specify the significant state of LH or RH dominant for each LS.