Paper Title
IPARO-Interactive Post-Mining Of Association Rules Using Ontologies
Abstract
In Data Mining, the usefulness of association rules is strongly narrow by the huge amount of delivered rules. To
overcome this drawback, we can make use of several techniques such as redundancy reduction, itemset concise
representations, and postprocessing. However, as per statistical information, these methods do not assurance that the
extracted rules are interesting for the user. Thus, it is difficult to the decision maker and requires essential help with an
efficient postprocessing step in order to reduce the number of rules. This paper proposes a new interactive approach to prune
and filter discovered rules. First, we propose the use of Ontologies to improve the integration of user knowledge in the
postprocessing task. Second, we propose the Rule Schema so that the user can express goals and expectations concerning the
association rules. And finally, an interactive framework is designed to assist the user throughout the analyzing task. Our new
approach is applied over voluminous sets of rules, by integrating domain expert knowledge in the postprocessing step, to
reduce the number of rules to several dozens or less. Moreover, the quality of the filtered rules was validated by the domain
expert at various points in the interactive process.