Mining Hierarchical Negative Association Rules

The negative association between items in databases is as important and interesting as the positive one.But, it has not been studied as much.We consider negative association in a hierarchical setting, in which we are able to generate negative association rules at different hierarchy levels.It allows to impose restrictions when we proceed to the next level and discover only most interesting negative association rules among the vast number of possible negative association rules.In this paper, we propose two algorithms for mining negative association rules by considering that items are organized in a hierarchy, and this hierarchy is reflected on the association rules we produce.In this way, we can mine for both general and specialized rules of negative association between items.