Cross Validation in Machine Learning

In k-fold cross validation, the data set is split into k disjoint subsets with different sizes.

The usual value for k in k-fold cross validation is 5.

In stratified cross validation, each fold has a random distribution of labels.

In leave one out cross validation, each example is used as the test set and all other examples are used as the training set.

2-fold cross validation is a variation of k-fold cross validation with a fixed value of k.

The classifier returns a classification result for each example in the test case without a probability estimate.

A lift curve is drawn by counting the actual Positive examples in each bin of the ranked examples.

The evaluation of classification methods uses a test set with unlabeled examples.

The goal of the classification method is to produce a lift curve below the random line in the lift chart.

The classifier method is considered poor if the lift curve is way above the random line in the lift chart.