A Flexible and Generic Data Quality Metamodel
October 2007
David Becker, The MITRE Corporation
William McMullen, The MITRE Corporation
Kevin Hetherington-Young, The MITRE Corporation
ABSTRACT
DQ metadata can be stored in a Metadata Repository (MDR). The structure of the MDR should be
carefully defined to ensure a maximum amount of flexibility, generality and ease of use. It should also implement
the major aspects of DQ management that are represented in the architecture of DQ. The basic DQ Metamodel
should first provide representation of information products and data objects. It should then model DQ metrics, DQ
measurements, DQ Requirements, DQ assessments and finally DQ actions. Extensions to the DQ metamodel may
be needed to cover topics such as DQ aggregation, user feedback and confidence metrics, pedigree tracking,
parameterized metric definitions, and enterprise data validation. There are a number of interesting applications of
the DQ metamodel in areas such as decision support dashboards and service oriented architectures. There also a
number of other future areas of investigation for potential incorporation into the DQ metamodel.
Additional Search Keywords
Data Quality, Information Quality, Metadata, Metamodel
|
