The Classification of Valid and Invalid Beats of Three-Dimensional Nystagmus Eye Movement Signals Using Machine Learning Methods

Figure 6

Classification results of the different methods used (nn is the nearest neighbour) as accuracies (as percentages) with dataset variations: (1) manual selection and no cleaning of rejected nystagmus candidates, (2) manual selection and with cleaning of rejected candidates, (3) automatic selection and no cleaning, (4) automatic selection with cleaning, (5) both manual and automatic selection and no cleaning, and (6) both manual and automatic selection with cleaning.