Database design process is a knowledge intensive task that requires expertise, practical experience, and judgment. It is not surprising, therefore, that over the last few years many research prototype database design expert systems have been reported in the literature. This paper is a survey of such tools. These tools are compared with respect to four major aspects: database design support, tool flexibility, expert system features, and implementation characteristics. This study reveals that, in general, there is lack of 1) support for all the phases of the design, 2) support for group database design, 3) graphic support, 4) empirical verification of effectiveness of the tools, 5) long-term maintenance of the tool and database schemata, and 6) specialized knowledge representation schemes, inference, and learning techniques.
Unique complements and decompositions of database schemata
