Efficient Ontology Meta-Matching Based on Interpolation Model Assisted Evolutionary Algorithm

Ontology is the kernel technique of Semantic Web (SW), which models domain knowledge in a formal and machine-understandable way. To ensure different ontologies’ communications, cutting-edge technology to determine heterogeneous entity mappings through ontology matching process. During this procedure, it utmost importance integrate similarity measures distinguish correspondence. The way find most appropriate aggregating weights enhance alignment’s quality called meta-matching problem, recently, Evolutionary Algorithm (EA) has become great methodology addressing it. Classic EA-based evaluates each individual traversing reference alignment, increases computational complexity algorithm’s running time. For overcoming drawback, an Interpolation Model assisted EA (EA-IM) proposed, introduces IM predict fitness value newly generated individual. In particular, we first divide feasible region into several uniform sub-regions using lattice design method, then precisely evaluate Interpolating Individuals (INIDs). On basis, constructed for new forecast its value, with help neighborhood. testing EA-IM’s performance, use Alignment Evaluation Initiative (OAEI) Benchmark experiment final results show that EA-IM capable improving EA’s searching efficiency without sacrificing solution’s quality, f-measure values are better than OAEI’s participants.