| Summary: | Database designers usually spend much time integrating database views created by
different users. This is because different users have different perceptions of the real world.
Therefore, conflicts exist between database views. The most common conflicts found are
naming and structural ones.
Wagner (1989) suggests four characteristics (name, meaning, construct and context)
that can be used to identify and distinguish database elements. He also provides a complete
solution for the inter-view conflicts. Finally, he derives some heuristic rules that are used to
determine similar database elements. However, Wagner's approach does not provide
database designers with an efficient conflict analysis procedure. Moreover, no direction is
given to database designers for the comparison of different database elements. So, database
designers have to compare unnecessary elements.
The objective of this research is to lessen the requirement of the view integration
system to interact with database designers during a conflict analysis. A more efficient
method of conflict analysis is outlined which compares elements in the following order: 1)
entities, 2) relatedness, 3) relationships and 4) attributes. Fuzzy logic is also used to provide
database designers with some numerical analysis of the degree of similarity between entities.
This concept is used at the stage of entities and relatedness identification. The improved
view integration system reduces much of the database designers' workload by comparing
fewer elements during the conflict analysis. === Business, Sauder School of === Management Information Systems, Division of === Graduate
|
|---|
