Short Time Association Rule Mining Algorithm

Many algorithms have been proposed to solve the problem of mining frequent itemset. The resulting frequent itemsets represent the global frequent patterns. This global output doesn´t provide any information about the distribution of the frequent patterns on the database. This missing information can produce inaccurate decisions or prediction when the output frequent itemsets are used in decision support or prediction systems, specially, when the input database is non-uniformly distributed. In this work, we shall introduce a technique to calculate, store, and display frequent itemsets´ distributions in the database. The proposed technique is called short time association rule mining (ST-ARM).