Skip to main content

Table 2 Groupwise average accuracies of various feature extraction methods combined with various classifiers applied to classify sets A, B, C, D, and E (standard deviations are noted in parentheses).

From: Combination of EEG Complexity and Spectral Analysis for Epilepsy Diagnosis and Seizure Detection

Classifier Feature Selection A B C D E Accuracy
LLS All features 94.75 (3.6) 88.75 (5.0) 51.00 (8.6) 60.00 (8.0) 92.25 (3.6) 77.95 (1.7)
LDA All features 96.00 (2.9) 88.25 (5.3) 51.00 (8.6) 62.00 (8.1) 94.25 (4.1) 78.30 (1.7)
BP All features 95.75 (3.1) 91.50 (3.8) 59.25 (8.8) 63.75 (9.3) 94.00 (2.7) 80.85 (1.5)
  PCA 94.75 (4.9) 92.50 (3.3) 63.25 (7.0) 70.50 (9.5) 93.00 (2.8) 82.80 (2.5)
LISVM All features 94.75 (6.5) 92.00 (4.4) 59.50 (5.8) 64.50 (6.5) 95.25 (1.4) 81.20 (1.4)
  PCA 94.50 (4.4) 91.75 (4.7) 62.50 (7.7) 63.00 (9.1) 94.50 (3.5) 81.25 (2.2)
RBFSVM All features 97.00 (4.4) 93.00 (3.5) 71.75 (6.5) 70.00 (8.7) 96.00 (3.4) 85.55 (2.0)
  PCA 94.25 (7.3) 92.50 (3.7) 75.75 (7.4) 72.00 (6.7) 95.00 (3.3) 85.90 (2.4)