Commonsense knowledge support in database design expert systems

• Main Author: Ding, Jie
• Format: Others
• Language: English
• Published: 2009
• Online Access: http://hdl.handle.net/2429/5262

Conceptual database design is the most critical and difficult phase in designing a database centered application. It usually requires database design experts which are hard to find and expensive. There has been some effort in building expert systems, including the View Creation System (VCS), to support this design process. However, all of these systems lack commonsense knowledge human experts have and therefore can not provide effective support to the user. A prototype commonsense knowledge base, the Commonsense Business Reasoner (CBR), has been built for the VCS. But it is not filly implemented and not connected to the VCS. The usefulness of commonsense knowledge in a database design expert system has not been studied. In this paper, a new domain of relevance ontology was proposed to store domain of relevance for commonsense knowledge. Combined with the Naive Semantics ontology used in the CBR, a new commonsense knowledge base structure was built and implemented. This was integrated into the original VCS and can provide interactive commonsense knowledge support during the design process. Other improvements were also done to the VCS to make it more user friendly. A fill scale empirical study was conducted to test the effectiveness of the commonsense module. The results indicated subjects perceived the system with commonsense knowledge easier to use and finished the task in less time. However, there is no statistically significant difference in the design quality. Explanations to these results are discussed as well as future research directions. === Business, Sauder School of === Graduate