Mining associations with the collective strength approach

The large itemset model has been proposed in the literature for finding ass ociations in a large database of sales transactions. A different method for evaluating and finding itemsets referred to as strongly collective itemset s is proposed. We propose a criterion stressing the importance of the actua l correlation of the items with one another rather than their absolute leve l of presence. Previous techniques for finding correlated itemsets are not necessarily applicable to very large databases. We provide an algorithm whi ch provides very good computational efficiency, while maintaining statistic al robustness. The fact that this algorithm relies on relative measures rat her than absolute measures such as support also implies that the method can be applied to find association rules in data sets in which items may appea r in a sizeable percentage of the transactions (dense data sets), data sets in which the items have varying density, or even negative association rule s.