Discovering, Maintaining, and Using Conceptual-Relational Mappings

Yuan An
College of Information Science and Technology
Drexel University

Abstract:
Conceptual-relational mappings between conceptual models
and relational schemas have been used increasingly to achieve
interoperability or overcome impedance mismatch in modern
data-centric applications.

In this talk, I will describe our study on discovering and maintaining
conceptual-relational mappings and  using the
mappings to facilitate the discovery of direct mappings between relational
schemas. We developed a semi-automatic tool, MAPONTO, which takes as input a
database schema, a conceptual model (CM), and a set of simple correspondences
between atomic elements in the schema and the CM. It then
generates a set of logical formulas  that define a mapping from the
schema to the CM.  To deal with CM/schema changes, we proposed a
round-trip engineering solution which essentially synchronizes models
by keeping them consistent for maintaining conceptual-relational mappings.
We also extended MAPONTO with a semantic
approach to finding schema mapping expressions. The approach leverages
the semantics of schemas expressed in terms of conceptual-relational
mappings. Our
evaluation results show that MAPONTO is effective in discovering and maintaining
conceptual-relational mappings and it outperforms the traditional
mapping techniques for building complex schema mapping expressions in
terms of both recall and precision. Finally, I will discuss our experience
in developing the MAPONTO tool and present several future research
questions.