Ontology inconsistent problem on Model-based Diagnosis

With the development of semantic web, ontology inconsistent problem has been one of the most important research directions of ontology alliance. Based on the difference of research method and problem domain, there are a serial of solutions, among which, Model-based Diagnosis has been widely used in the domain of constructive knowledge description. To obtain efficient methods of diagnosis,local diagnosis technology uses information about the structural knowledge of domain to contract problem reasoning space. On the basis of local diagnosis technology, this paper proposes an “innocent diagnosis” method, which ulteriorly contract problem reasoning space through deleting the truth assumables proved by new observation, presents an E_DAG figure representation method of inconsistent problem and shows its working principle through the MadCows ontology example at the end.