Abstract: Data mining is becoming a popular research topic with its frequent applications in online e-business, web click stream analysis and cross marketing. Mining high utility itemsets from a transactional database is concerned with the discovery of itemsets with high utilities like profits or gains. Efficient discovery of the frequent and useful itemsets in huge datasets is a crucial task in data mining. In the recent years, many methods have been proposed for generating high utility patterns. Owing to this, there are few problems such as, if the minimum utility threshold value is set too low, huge amount of itemsets are generated. But, if the minimum utility threshold is set too high, very few or no high utility itemsets will be generated. In high utility itemset mining, the profit values or utility value for every item and the number of units of each item is taken into consideration. We hereby present the study of issues related to the different structures used and algorithms for mining the high utility itemsets.

Keywords: Data mining; frequent itemset; high utility itemset; transactional database.