Distributed ontology based information retrieval using semantic web

Chun Zhang

Shantou Radio and TV University,Shantou, Guangdong, 515041, China

In recent years, user demand for integrated searches over several independently operating semantic web systems have been increasing rapidly. Integrated semantic searches enable more meaningful results to be generated because information with similar meanings in diverse areas and domains is likely to be used for inference. However, it is not easy to integrate physically independent, distributed, and heterogeneous database systems to provide a single, integrated semantic web system to end-users. In this paper, we propose a novel system that integrates heterogeneous semantic web systems based on schema mapping. The user can generate only one SPARQL query using the integrated schema without the necessity of checking the schemas of individual systems each time thereby reducing additional costs to generate queries for individual systems. Furthermore, the user is not required to collect individual query results manually after performing a query and additional costs for establishing systems can be reduced because no change in existing system structures is required. If currently established systems are expanded by adding the schema structures of other ontology systems, the cost to establish another integrated retrieval system can be saved. To evaluate the effectiveness of our approach, we have implemented a prototype that integrates two national information retrieval systems.