A Fast Algorithm to Maintain the Discovered High-Utility Itemsets with Modified Records

Abstract

High-utility item sets mining (HUIM) is a critical issue to concern the quality and profit factors to mine high utility item sets (HUIs). Many algorithms have been proposed to efficiently mine HUIs from a static database. The discovered HUIs may become invalid or new HUIs may arise when transactions are modified in the original database. In the past, prelarge concept was proposed to efficiently maintain and update the discovered information of association-rule mining. In this paper, a maintenance algorithm for transaction modification based on a designed prelarge strategy is presented to efficiently maintain and update the discovered HUIs. Based on the designed algorithm, the original database is unnecessary to be rescanned until the accumulative total utility of the modified transactions achieves the designed safety bound, which can greatly reduce the computations of multiple database scans compared to the previous batch-mode approaches.