A new temporal measure for interesting frequent itemset mining

Frequent itemset mining assists the data miner in searching for strongly associated items (and transactions) in large transaction databases. Since the number of frequent itemsets is usually extremely large and unmanageable for a human user, methods for mining interesting rules have been proposed to define meaningful and summarized representations of them. Furthermore, many measures have been proposed in the literature to determine the interestingness of the rule. In this paper, we introduce a new temporal measure for interesting frequent itemset mining. This measure is based on the idea that interesting frequent itemsets are mainly covered by many recent transactions. This measure reduces the cost of searching for frequent itemsets by minimizing the search interval. Furthermore, this measure can be used to improve the search strategy implemented by the Apriori algorithm.