BioMed Research International / 2016 / Article / Tab 2

Research Article

Brain-Computer Interface for Control of Wheelchair Using Fuzzy Neural Networks

Table 2

Comparison of classification results.

MethodCorrectly classified instancesIncorrectly classified instancesMean absolute error Root mean squared error

Linear logistic regression model96%4%0.02140.1265
SVM (polynomial kernel)100%00.240.3162
SVM (RBF kernel)74%26%0.25680.3404
SVM (PUK kernel)96%4%0.24240.32
MLP (NN) (5 hidden neurons)88%12%0.07240.1586
MLP (NN) (6 hidden neurons)100%00.0480.0958
Naïve Bayesian94%6%0.0240.1549
Random Tree74%26%0.1040.3225
Random Forest98%2%0.12150.179
FNN (6 hidden neurons)100%01.8230.257986