This program computes the features for the eye movement identification problem and store them in ARFF format, which can be processed by Weka. Details of the feature construction algorithm can be found in the following paper:
Nguyen Viet Cuong, Vu Dinh, Lam Si Tung Ho. Mel-frequency Cepstral Coefficients for Eye Movement Identification. IEEE International Conference on Tools with Artificial Intelligence (ICTAI 2012).
This program uses the voicebox toolbox for MATLAB at: http://www.ee.ic.ac.uk/hp/staff/dmb/voicebox/voicebox.html
Input to the program is a file named 'data.emd', which stores eye movement data in EMD1 format. This is a small subset of the dataset A in the EMVIC 2012 Competition. A description of the EMD1 format can be found at: http://www.emvic.org/