Methods. Association rule mining was first introduced in 1993 by [1]. An association rule is of the form x--y, where x and y are sets of items and xny= 0. The main task of association rule mining is to extract from the database all rules that satisfy the minimum support and the minimum confidence constraints specified by the user. Given a database with records each with a class label, classification rule mining aims to extracta set of concise rules for each class to form an accurate classifier. The main difference between classification rules and association rules is that association rules do not have predetermined target or consequent, whereas classification rules have a pre-determined target, which is the class.

The integration of association rule mining with classification rule mining to mine rules whose consequents are restricted to the classification class attribute was first introduced by [2]. Our approach is similar to thiswork, but our proposed algorithm is different from the association algorithm and [2], because we use odds ratio instead of support and confidence in the measure of interestingness.