Page 237 - DCAP603_DATAWARE_HOUSING_AND_DATAMINING

Unit 12: Metadata and Warehouse Quality

notes
figure 12.2: the Data Warehouse architecture Meta Model

The proposed metamodel (i.e. the topmost layer in Figure 12.2) provides a notation for data
warehouse generic entities, such as schema or agent, including the business perspective. Each
box shown in Figure 12.2 is decomposed into more detailed data warehouse objects in the
metamodel. This metamodel is instantiated with the metadata of the data warehouse (i.e. the
second layer in figure 12.2), e.g. relational schema definitions or the description of the conceptual
data warehouse model. The lowest layer in Figure 12.2 represents the real world where the actual
data reside: in this level the metadata are instantiated with data instances, e.g. the tuples of a
relation or the objects of the real world which are represented by the entities of the conceptual
model.

12.3.1 Quality Meta Model

Each object in the three levels and perspectives of the architectural framework can be subject to
quality measurement. Since quality management plays an important role in data warehouses,
we have incorporated it into our metamodeling approach. Thus, the quality model is part of the
metadata repository, and quality information is explicitly linked with architectural objects. This
way, stakeholders can represent their quality goals explicitly in the metadata repository, while,
at the same time, the relationship between the measurable architecture objects and the quality
values is retained.

The DWQ quality metamodel is based on the Goal-Question-Metric approach (GQM) of originally
developed for software quality management. In GQM, the high-level user requirements are
modeled as goals. Quality metrics are values which express some measured property of the
object. The relationship between goals and metrics is established through quality questions.

LoveLy professionaL university 231