Generation and Matching of Ontology Data for the Semantic Web in a Peer-to-peer Framework

Abstract. The abundance of ontology data is very crucial to the emerg-ing semantic web. This paper proposes a framework that supports the generation of ontology data in a peer-to-peer environment. It not only enables users to convert existing structured data to ontology data aligned with given ontology schemas, but also allows them to publish new on-tology data with ease. Besides ontology data generation, the common issue of data overlapping over the peers is addressed by the process of ontology data matching in the framework. This process helps turn the implicitly related data among the peers caused by overlapping into ex-plicitly interlinked ontology data, which increases the overall quality of the ontology data. To improve matching accuracy, we explore ontology related features in the matching process. Experiments show that adding these features achieves better overall performance than using traditional features only.