“…The classification for an unclassified record is based on comparing it with similar records in the training dataset. The distance between the new record and the closest class is measured by a distance function such as Euclidean, Manhattan or Minkowski for continuous variables and Hamming distance for categorical variables [52].…”