By looking at the mean of each class for each specific feature it is possible to observe that the collected data has a particular mean for each of the eye movements that were sampled with the pre-processing circuit of Figure \ref{149602}, besides the variance of ARV and RMS for each class is small. That helps to determine that the observation is not far away from the mean value of all observations.  At first glance, it may seem that the select features could help to classify better the eye movements of the users.  To visualize better the features and the classes it is possible to draw and analyze the scatter plot (Figure \ref{984894}) of the label data of the 20 sampled registers. This also helps to visualize if the classes are linearly separable and select the proper classification technique to process the information of the electrooculography signal.