These are two-sample tests for categorical data utilizing similarity information among the categories. They are useful when there is underlying structure on the categories.
