Pushing Constraint in Association Rules Mining: an Ontology-Based Approach

This paper proposes an integrated framework for the extraction of constraint-based multi-level association rules with the aid of an ontology. The latter, that represents an enriched taxonomy, is used to describe the application domain by means of data properties. Defining or updating these properties is a simple task and does not imply changing the items hierarchy, or the implementation level of our framework. The system enables the definition of domain-specific constraints by using the ontology to filter the instances used in the association rule mining process. This can improve the quality of the extracted associations rules and make them more interesting and easy to understand. We describe our framework, also including examples of queries based on real-data.