A Survey on Discovery of High Utility Itemset Mining From Transactional Databases
Publication Date : 31/03/2016
Utility mining is a recent topic in the ﬁeld of data mining. To ﬁnd out the highest utility itemsets, in terms of proﬁt, quantity, cost or other user preferences is the main goal of utility Mining. The high utility itemset (HUI) mining using a transactional database is to explore itemsets that have utility above a user speciﬁed threshold. High utility itemset mining is an addition to frequent itemset mining, which consists of itemsets discovery that are frequently bought together by customers. Rare items considered as high utility itemsets in many real-life applications. For several decision making domains like business transactions, medical, security, fraudulent transactions, retail communities etc, rare itemsets give helpful information. For example, in a supermarket, customers purchase microwave ovens or frying pans infrequently as compared to bread, washing powder, soap. But the later transactions yield less proﬁt compared to former transactions for the supermarket. Thus the high proﬁt rare itemsets are found to be very helpful in many application areas. The performance of utility mining can be further enhanced with compact and lossless representation of high utility itemsets. This study analyzes various methods for high utility itemset mining using transactional databases.
No. of Downloads :