ISBN: 978-981-11-3671-9 DOI: 10.18178/wcse.2017.06.121
Is Meditation Measurable?
Abstract— This paper discusses how to utilize entropy to brain states classification based on EEG recordings. We
use 3 different classification models, one is tree based, one is distance based, and the other is probability based. We
discuss the suitability of 2 types of entropy: approximate entropy, and sample entropy in different circumstances. With
proper choice of parameters, misclassification rate is less than 5%. Sample entropy is a good tool to extract information
from EEG data.
Index Terms— meditation; machine learning; electroencephalogram (EEG); entropy; classification
Hong Lin
Department of Computer Science and Engineering Technology, University of Houston-Downtown, US
Yuezhe Li
Center for Cell Analysis and Modeling, University of Connecticut Health Center, US
Cite: Hong Lin, Yuezhe Li, "Is Meditation Measurable?," Proceedings of 2017 the 7th International Workshop on Computer Science and Engineering, pp. 696-700, Beijing, 25-27 June, 2017.