The classification errorEi of an individual model
i measures the number of records incorrectly classified
(false positives plus
false
negatives) and is evaluated by the formula:
where f is the number of records incorrectly classified, and
n is the total number of records.