Verification of Data in Staging Schema, Mapping Between Source Tables and Materialized Views

With the proliferation of international standards for biomedical databases, the need for integration of small heterogeneous databases into large data warehouses highlights issues of consistency and integrity. In this paper, we discuss methods for providing semi-autonomous data integration by focusing on efficient design and effective mapping strategies. In order to upload and integrates data from various small to medium sized data repositories, an intermediate staging facility is employed to temporarily store data before it is validated to ensure accurate and useful integration into the data warehouse. We propose a mechanism that semi-automates the integration process that includes not only new data, but legacy data as well. We expand the notation of a DBMS to include the management of the data transfer process. The DBMS now must perform the task of data mapping in the form of value correspondences by using a staging schema mapping procedure