Conceptual Hierarchies Matching: An Approach Based on Discovery of Implication Rules Between Concepts

Most research works about ontology or schema matching are based on symmetric similarity measures. By transposing the association rules paradigm, we propose to use asymmetric measures in order to enhance matching. We suggest an extensional and asymmetric matching method based on the discovery of significant implications between concepts described in textual documents. We use a probabilistic model of deviation from independence, named implication intensity. Our method is divided into two consecutive stages: (1) the extraction in documents of relevant terms for each concept; (2) the discovery of significant implications between the concepts. Our method is tested on two benchmarks. The results show that some relevant relations, ignored by a similarity-based matching, can be found thanks to our approach.