Matching relevant ontology data for integration is vitally important as the amount of ontology data increases along with the evolving Semantic web, in which data are published from different individuals or organizations in a decentralized environment. For any domain that has developed a suitable ontology, its ontology annotated data (or simply ontology data) from different sources often overlaps and needs to be integrated. The purpose of this paper is to develop intelligent web ontology data matching method and framework for data integration.
This paper develops an intelligent matching method to solve the issue of ontology data matching. Based on the matching method, it also proposes a flexible peer‐to‐peer framework to address the issue of ontology data integration in a distributed Semantic web environment.
The proposed matching method is different from existing data matching or merging methods applied to data warehouse in that it employs a machine learning approach and more similarity measurements by exploring ontology features.
The proposed method and framework will be further tested for some more complicated real cases in the future.
The experiments show that this proposed intelligent matching method increases ontology data matching accuracy.
