A Theory for Multiresolution Signal Decomposition: The Wavelet Representation
IEEE Transactions on Pattern Analysis and Machine Intelligence
Wrappers for feature subset selection
Artificial Intelligence - Special issue on relevance
An introduction to variable and feature selection
The Journal of Machine Learning Research
Pattern Classification (2nd Edition)
Pattern Classification (2nd Edition)
Introduction to Machine Learning (Adaptive Computation and Machine Learning)
Introduction to Machine Learning (Adaptive Computation and Machine Learning)
Hi-index | 0.00 |
The objective of this study is to analyze and classify evoked potentials obtained from familiar and unfamiliar face experiment. EEG signals were recorded from 26 volunteers. Multi-resolution analysis was used as a tool for signal approximation and modeling. A custom scaling-wavelet function pair and their bi-orthogonal complements were built by resembling the waveform of the scaling function to the excitatory post-synaptic potential. In order to distinguish the familiar-unfamiliar face evoked potentials, a Fisher's linear classifier was used with discriminative approximation coefficients obtained from active electrodes which are selected by the wrapper method. The algorithm was also executed with spline, Daubechies, Symlet and Coiflet wavelets for comparison. The classification performance of proposed wavelet is the first among the other wavelets with 69.7% accuracy and it is also first in the total number of highest success of individual subjects with 31% of the subjects which is double of the result of the second wavelet in the rank.