Association rule learning (Agrawal, Imieliski, & Swami, 1993, 1996) is a descriptive data mining technique which attempts to discover interesting relations between variables in large databases using unsupervised learning. This technique is tackled with Apriori (Agrawal et al., 1993). This algorithm obtains association rule with the form X ? Y, where X and Yare itemsets.The set of all items in the domain of investigation,consisting of a set of transactions. Apriori considers that transactions correspond to training examples of a data set, an item is a binary feature, and itemsets are conjunctions of features. The quality of an association rule obtained by Apriori are defined by its confidence and support, i.e. Confidence of a rule is the conditional probability of Y given X: p(YjX), and Support of a rule is an estimate of the probability of itemset X [ Y: p(X Y).