Click here to load whole tree
NITAAI-Veda.nyf > Soul Science God Philosophy > Science and Spiritual Quest > Section 4 Towards a New Biology > MATHEMATICAL TECHNIQUES FOR STUDY OF EEG > 3.3 Four-category problem

3.3 Four-category problem

 

Classification of data into shallow meditation, deep meditation, pre-meditation and post-meditation states are considered. The obtained accuracies for the back propagation network with 21 hidden units, RBF networks with 140 hidden units and support vector machine with RBF kernel are shown in Table 3. The highest classification accuracy (99.38%) is obtained for the support vector machine with RBF kernel.

 

Features Considered  KNNC  MLP    RBF    SVM

Mean, Variance        56.87  92.50  93.75  94.37

Fractal    Dimension (FD),     Complexity Measure (CM)         73.75  96.88  96.88  97.50

Mean,      Variance, FD, CM        69.37  96.25  96.25  97.50

Frequency   Domain Features        78.12  97.50  98.12  98.75

Mean,      Variance, FD,        CM        & Frequency    domain features 68.75  98.12  98.12  99.38

 

Table 3: Testing Accuracy Table for EEG Data (Four-category case)