Automatic Epileptic Seizure Detection Using Scalp EEG and Advanced Artificial Intelligence Techniques

Table 9

Cross-validation results from top five uncorrelated features from five regions.

Classifiers	80% holdout: 100 repetitions		Cross-validation, 5-fold, 1 repetition	Cross-validation, 5-fold, 100 repetitions
	Mean error	SD	Mean error	Mean error	SD
LDC	0.1690	0.0419	0.1696	0.1675	0.0120
QDC	0.1493	0.0449	0.1462	0.1509	0.0088
UDC	0.2926	0.0440	0.2836	0.2940	0.0037
POLYC	0.1690	0.0419	0.1871	0.1709	0.0091
LOGLC	0.1734	0.0413	0.1696	0.1648	0.0120
KNNC	0.1203	0.0339	0.0936	0.1135	0.0101
TREEC	0.1835	0.0460	0.1988	0.1784	0.0202
PARZENC	0.1328	0.0433	0.1316	0.1325	0.0146
SVC	0.1460	0.0378	0.1316	0.1411	0.0101