A Semantic Resource Based Approach for Star Schemas Matching

Abstract

The hybrid approach is widely used in constructing data warehouse (DW) schemas. It relies on a complex process for matching two sets of multidimensional star schemas: schemas built from business requirements (BR-Star schemas) and schemas constructed on the organization data source (DS-Star schemas). Using a semantic resource during this matching helps solving heterogeneity problems. This paper suggests a semiautomatic approach for the construction of approved star schemas by matching DS-Star schemas with BR-Stars, and by using WordNet as a semantic resource for solving heterogeneity issues. This approach consists of three steps: i) Match DS-Stars with BR-Stars; ii) Involve the DW designer for approval of the matching process; and then iii) Generate approved star schemas. We have defined two Boolean functions and four semantic metrics for the matching process. We have developed a software prototype for testing our approach.

An Automatic Schema-Instance Approach for Merging Multidimensional Data Warehouses

“…The selection metric of bipartite graph helps determine the mapping choice and define rules aiming at preserving the partial orders of the hierarchies at mapping time. Another approach matches a set of star schemata generated from both business requirements and data sources [5]. Semantic similarity helps find the matched facts and dimension names.…”