4.2.1.2. Distance-Based Information Loss (DBIL). The second measure
is the distance-based information loss (DBIL)[18] a direct comparison
of values between datasets X (original) and X0 (masked). Then,
the difference between the original dataset X and the masked dataset
X0 can be defined as the sum of the distances between each original
register rX and its corresponding protected register rX0 (see Eq.
(13)), where the distance between two registers is the sum of the
distances between each original category and its corresponend
protected category for all variables Vi (see Eq. (14)).