Using an interest ontology for improved support in rule mining

This paper describes the use of a concept hierarchy for improving the results of association rule mining. Given a large set of tuples with demographic information and personal interest information, association rules can be derived, that associate ages and gender with interests. However, there are two problems. Some data sets axe too sparse for coming up with rules with high support. Secondly, some data sets with abstract interests do not represent the actual interests well. To overcome these problems, we are preprocessing the data tuples using an ontology of interests. Thus, interests within tuples that axe very specific axe replaced by more general interests retrieved from the interest ontology. This results in many more tuples at a more general level. Feeding those tuples to an association rule miner results in rules that have better support and that better represent the reality.(3)