P: A Model for Semantic Equivalence Discovery for Harmonizing Master Data

Ken Evans — Wed, 11 Nov 2009 10:38:49 GMT

A Model for Semantic Equivalence Discovery for Harmonizing Master Data

Baba Piprani MetaGlobal Systems, Canada
Abstract. IT projects often face the challenge of harmonizing metadata and data so as to have a “single” version of the truth. Determining equivalency of multiple data instances against the given type, or set of types, is mandatory in establishing master data legitimacy in a data set that contains multiple incarnations of instances belonging to the same semantic data record . The results of a real-life application define how measuring criteria and equivalence path determination were established via a set of “probes” in conjunction with a score-card approach. There is a need for a suite of supporting models to help determine master data equivalency towards entity resolution—including mapping models, transform models, selection models, match models, an audit and control model, a scorecard model, a rating model. An ORM schema defines the set of supporting models along with their incarnation into an attribute based model as implemented in an RDBMS.Keywords: Entity Resolution, Master Data, Semantic Equivalence, semantic interoperability, data equivalencyLNCS 5872, p. 649 ff.

ORM 2009 : MDM

P: A Model for Semantic Equivalence Discovery for Harmonizing Master Data