For the interoperability of heterogeneous information management systems, schema mapping approaches have been used to build a unified view. The schema mapping approach offers full transparencies and is very powerful from the user’s point of view. However, the traditional mapping approach needs to be strengthened for information management systems that have non-traditional data types, no schema, or incompatible schemas. We have incorporated numerous concepts and constructs associated with the knowledge and object-oriented paradigm such as abstract views with a set of procedures, encapsulation, inheritance and class composition hierarchies to resolve the above problem. This extension also accommodates the ability to determine and explicitly represent the semantics in the schema. Additionally, we have outlined a query processing method using the unified view. We are currently developing a prototype to support seamless access to structured data and unstructured data managed by different information management systems.
