Classification with a Neutral Zone
Daniel R. Jeske · Wiley StatsRef: Statistics Reference Online · 2021
Abstract Statistical classifiers are of limited use when one or both of the conditional misclassification rates is unacceptably high. Incorporating a neutral zone region into the classifier provides a mechanism to defer a classification decision for ambiguous cases. By using a neutral zone region the conditional misclassification rates can be controlled and the usefulness of the classifier is greatly enhanced. Neutral zone classifiers can also be incorporated into the first stage of a two‐stage classification procedure to reduce the cost of collecting expensive features.