Ontology Mapping Strategies And Their Combination Based On Semantics

As the number of ontologies keeps growing, ontology sharing and reuse become urgent problems. Mapping between different ontologies is the key to solve the problems of ontology communication and interoperation.Firstly, the research background of this topic is introduced, and development ontology mapping technique is summarized.Secondly, aiming to solve the problem that traditional similarity computing strategies ignore the semantic environment, the concept of "conditional information quantity" is proposed, and new mapping strategies are designed based on this. The new strategies combine the ontology characteristics (like individuals, comments, name and structure) with their semantic environments, emphasizing the more semantically valuable information, and filter noise information.Thirdly, a semantics-based multi-strategy combination method is designed, aiming to solve the problem that traditional combination methods do not sufficiently use semantic information in every strategy. The new combination method sufficiently uses the semantic meaning and semantic relation among them, overcoming the shortages of strategies. And it also arranges adequate position for every strategy, so absence or repetition of some information will not have much impact on the whole system.Finally, SNAXMapping2 has been designed and implemented according to the works above. Experiment result shows that this system can improve the precision while maintaining recall, thus gets a better mapping result.