[[alternative]]An Efficient Approach for Maintaining Association Rules based on Adjusting FP-tree Structure

[[abstract]]In this thesis, the issue of mining and maintaining association rules in a large database of customer transactions is studied. The maintenance of association rules can be mapped into the problem of maintaining frequent itemsets in the database. Because the mining of association rules is time-consuming, we need an efficient approach to maintain the frequent itemsets when the database is updated. In this study, a general incremental updating technique is proposed for maintaining the frequent itemsets discovered in a database in the cases including insertion, deletion, and modification of transactions in the database. An efficient algorithm, called AFPIM (Adjusting FP-tree for Incremental Mining), is designed based on adjusting FP-tree structures. Our approach uses a FP-tree structure to store the compact information of transactions involving frequent and pre-frequent items in the original database. In most cases, the new FP-tree structure of the updated database can be obtained by adjusting FP-tree of the original database according to the changed transactions without needing to rescan the original database. Experimental results show that AFPIM outperforms the existing algorithms in terms of the execution time.