BMC Medical Informatics and Decision Making

Table 5 Results of all binary classification models with their respective cross-validation and their accuracy of the unbalanced database

From: A comparative study of CNN-capsule-net, CNN-transformer encoder, and Traditional machine learning algorithms to classify epileptic seizure

			Epileptic	Without Epileptic
	Accuracy [%]	Cross Validation [%]	Precision [%]	Precision [%]
ETC	97.87	97.80 ± 0.5	96.60	98.20
RFC	98.13	98.00 ± 1.0	96.30	98.60
GB	97.43	97.40 ± 2.0	97.70	97.40
DTC	94.09	94.00 ± 1.0	85.70	95.60
MLP	98.13	98.00 ± 0.5	96.80	98.10
KNN	94.78	95.00 ± 0.4	99.10	94.00
SGD	83.87	83.90 ± 1.0	95.70	83.80
SVM	98.30	98.00 ± 0.3	96.70	98.70
CNN+Fully	99.61	99.00 ± 1.0	99.80	99.90
CNN+Capsule-Net	99.35	99.30 ± 1.0	98.10	99.70
CNN+TF	99.83	99.80 ± 0.3	99.80	99.80
CNN+Tf+FULLY	99.74	99.80 ± 1.0	99.60	99.80
CNN+TF+Capsule-Net	99.65	99.70 ± 1.0	99.60	99.70

Back to article page

ISSN: 1472-6947

Contact us

General enquiries: journalsubmissions@springernature.com