An Efficient Algorithm for High Utility Mining

Mining high utility itemsets from a transactional database refers to the discovery of itemsets with high utility like profits. Efficient discovery of high utility itemsets from a transactional database is a crucial and difficult task. Given a transactional database and a profit table containing profit of each item present in the database, high utility itemsets can be discovered. In recent years, several approaches have been proposed for generating high utility itemsets, but they incur the problems of producing a large number of candidate itemsets for high utility itemsets. The situation may become worse when the database contains lots of long transactions or lots of unpromising items. In this paper, we propose HUP-Tree (High Utility Pattern Tree), for maintaining information of high utility itemsets, which is then mined to produce high utility itemsets. The proposed methods reduce the number of candidate sets efficiently and improve the mining performance in terms of execution time and space requirement. This approach is quite effective especially when transactions contain lots of unpromising items