Historical Background
In statistics or data mining, a typical task is to learn a
model from available data. Such a model may be a
regression model or a classifier. The problem with evaluating
such amodel is that it may demonstrate adequate
prediction capability on the training data, but might
fail to predict future unseen data. cross-validation is a
procedure for estimating the generalization performance
in this context. The idea for cross-validation originated
in the 1930s [6].