Maintenance of the Prelarge Trees for Record Deletion

The frequent pattern tree (FP-tree) is an efficient data structure for association-rule mining without generation of candidate itemsets. It, however, needed to process all transactions in a batch way. In addition to record insertion, record deletion is also commonly seen in real-application. In this paper, we propose the structure of prelarge trees for efficiently handling deletion of records based on the concept of pre-large itemsets. Due to the properties of pre-large concepts, the proposed approach does not need to rescan the original database until a number of records have been deleted. The proposed approach can thus achieve a good execution time for tree construction especially when a small number of records are deleted each time. Experimental results also show that the proposed approach has a good performance for incrementally handling deleted records. Key-Words: data mining, FP-tree, Prelarge-tree algorithm, pre-large itemsets, record deletion.