With the rapid development of Computer & Network, integration of heterogeneous data sources is becoming one of the hotspots in the research. Information integration system attempt to provide user with seamless and flexible access to information from multiple anonymous, distributed and heterogeneous data sources through a unified query interface. Information integration traditionally was dealing with the heterogeneities in structure and syntax. The heterogeneities in semantics are standing out while the data repositories are becoming large in size, dynamic, and widely distribution, in addition to the lack of the unified vocabulary among the data sources. In this paper, we propose a semantics-oriented information integration prototype system -- ontology-driven semantics-oriented information integration method, which introduced the ontology as the unified semantic vocabulary into implement the data sharing and interaction among the multiple data sources.
Firstly, the dissertation introduced the concept of ontology, the ontology mapping and the query technology. Next how to set up the mapping and how to build the ontology were introduced. A ontology-based semantic query method was introduced. This query statement based on the result of reasoning global ontology was decomposed to produce the query statements on different data sourced to realize the goal of unified visit. Then the ontology-based semantic-oriented query prototype system was designed, and the individual need of user was considered. Finally the implementation of prototype system is introduced, including the choice of key technology, the introduction of key interfaces, methods and system test.
Key Words：Ontology, Heterogeneous Data Integration, Sementic, Query
