Mining actionable combined patterns satisfied both utility and frequency criteria

Publication Type:
Thesis
Issue Date:
2016
Full metadata record
Files in This Item:
Filename Description Size
01front.pdf87.05 kB
Adobe PDF
02whole.pdf753.59 kB
Adobe PDF
In recent years, the importance of identifying actionable patterns has become increasingly recognized so that decision-support actions can be inspired by the resultant patterns. A typical shift is on identifying high utility rather than highly frequent patterns. Accordingly, High Utility ltemset (HUI) Mining methods have become quite popular as well as faster and more reliable than before. However, the current research focus has been on improving the efficiency while the coupling relationships between items are ignored. It is important to study item and itemset couplings inbuilt in the data. For example, the utility of one itemset might be lower than a user-specified threshold, whereas the utility may be larger when an additional itemset takes part in; and vice versa, an item's utility might be high until another one joins in. In this way, although some absolutely high utility itemsets can be discovered, it is sometimes easy to find out that many redundant itemsets sharing the same item are mined (e.g., if the utility of a diamond is high enough, all its supersets are proved to be HUIs). Such itemsets are not actionable, as sellers cannot make higher profit if marketing strategies are created on top of such findings. To this end, this thesis introduces a new framework for mining actionable high utility association rules, called Combined Utility-Association Rules (CUAR), which aims to find high utility and strongly associated itemset combinations which include item/itemset relations. The algorithm is proved to be efficient per experimental outcomes on both real and synthetic datasets.
Please use this identifier to cite or link to this item: