High utility itemset mining refers to mine itemsets having utility value greater than user specified threshold from a transactional database.It is an important data mining task and has a wide range of real life applications.Mining high utility itemset is a challenging task and several approaches has been proposed in the recent years.The proposed approaches use different search strategy,data structure,pruning technique to make the approaches efficient.Algorithms like HUIM-MMU and MHU-Growth overcome the limitation of using a single threshold for the whole database.However they still struggle with 'rare item problem'.The most efficient high utility itemset mining algorithm EFIM introduces two upper bounds named sub-tree utility and local utility.By combining ideas from the HUIM-MMU,MHU-Growth and EFIM,I introduce two algorithm named HUIM-MMSU and HUIM-IMMSU.HUIM-MMSU is a candidate generation and retest based algorithm,which relies on sorted downward closure(SDC)property.On the other hand,HUIM-IMMSU uses a tree-like data structure.Experiment result shows that the proposed two algorithms can effectively discover high utility itemsets in a transactional database. |