Maintaining Discovered Frequent Itemsets: Cases for ChangeableDatabase and Support

Mining frequent itemsets from large databases hasplayed an essential role in many data mining tasks. It is alsoimportant to maintain the discovered frequent itemsets for these datamining tasks when the database is updated. All algorithms proposed sofar for the maintenance of discovered frequent itemsets are onlyperformed with a fixed minimum support, which is the same as that used toobtain the discovered frequent itemsets. That is, users cannot changethe minimum support even if the new results are unsatisfactory to theusers. In this paper two new complementary algorithms, FMP(First Maintaining Process) and RMP (Repeated Maintaining Process), areproposed to maintain discovered frequent itemsets in the case that newtransaction data are added to a transaction database. Both algorithmsallow users to change the minimum support for the maintenance processes.FMP is used for the first maintaining process, and when the resultderived from the FMP is unsatisfactory, RMP will be performed repeatedlyuntil satisfactory results are obtained. The proposed algorithms re-usethe previous results to cut down the cost of maintenance. Extensiveexperiments have been conducted to assess the performance of thealgorithms. The experimental results show that the proposed algorithmsare very resultful compared with the previous mining and maintenancealgorithms for maintenance of discovered frequent itemsets.