ISENS: A System for Information Integration, Exploration, and Querying of Multi-Ontology Data Sources

Separate data sources on related domains of knowledge generally contain different but complementary information. There are queries that can only be answered with pieces of information from some of the separate sources. Thus, it is of considerable interest to enable query answering based on searching the information in an integrated collection of sources. However, independently developed and evolved data sources generally use different schemas to represent their data. This makes it difficult to search the sources in an integrated way. To address this problem, we have developed an end-to-end information integration system that is based on the Semantic Web technologies, algorithms for efficient source selection, and a Web-based user interface to construct queries and search multi-ontology data. We describe the system architecture, handling of information integration in a multi-ontology environment, and the user interface capabilities for ontology visualization, query construction, and presentation of results. Moreover, we have developed a multi-ontology real-world data set up and measured the performance of the different ISENS components on query answering that involves information integration from the different data sources. The results indicate that the source selection and the logical reasoning parts of query processing are dominated by the time to transfer data from the selected sources and to load the triples in the reasoner. However, the query answering time is often of the order of a second or less allowing ISENS to be used efficiently for such semantic applications.