Modern information sources are very complex. Schemas
describing scientific databases (e.g. PIR)
can contain over a hundred relational tables or thousands of XML elements.
The same is often true for many business databases. How to effectively understand and
query such complex schemas is a non-trivial challenge, and one that we address
in this project.
Information is seldom maintained at a single source, as a result, a user query
task typically involves accessing and integrating information
from multiple sources, whose relationships are described using schema mappings.
With the increasing schema complexity, schema mapping complexity increases correspondingly.
Another goal of this project is to investigate approaches for efficiently using and
managing practical schema mappings between complex schemas.