Ontology Translation by Ontology Merging and Automated Reasoning

Ontology translation is one of the most dicult problems that web-based agents must cope with. An ontology is a formal speci- cation of a vocabulary, including axioms relating the terms. Ontology translation is best thought of in terms of ontology merging. The merge of two related ontologies is obtained by taking the union of the terms and the axioms de ning them. We add bridging axioms to one merge of two related ontologies not only as \bridges" between terms in two related ontologies but also make this merge as a new ontology for further merging with other ontologies. Automated reasoning by an inference engine (OntoEngine) can be conducted in the merged ontology in either a demand-driven (backward-chaining) or data-driven (forward chaining) way to implement translation. We illustrate our method by describing its application as an online ontology translation system, OntoMerge, which translate the dataset as a DAML le represented in the source ontology to the dataset represented in the target ontology also as a DAML le and it is stable to various datasets and ontologies. As uniform internal representation, Web-PDDL, which is developed as Web version of PDDL, is used for representing merged ontologies and datasets for automated reasoning