The probably most common way of representing rules from data mining is in the
form of propositional statements. Using attribute value conditions, we get
rules such as 
. This representation is called propositional logic,
and is described thoroughly in [Tru92, GGH89, RN95]. An
equivalent representation is a decision tree, of which an example is
given in Figure 
. An animal is either ok or dangerous,
depending on whether it is a green snake, another animal with big teeth or
something else.
Figure: An example of a decision tree.
A program sentence such as IF (Color=Red) AND (Size=Big) THEN Type=Pelican is also a propositional representation, and have the same expressive power (or lack of such) as propositional logic and decision trees.