Research On Data Integration Under Integrity Constraints Based On Relational Model

The goal of data integration is to provides a unificated applicated surface for the users to visit the isomerism data sources, thus the users truly concentrate the attention on the specific results which they wants, but does not need to care about how obtains these results. This article has carried on the generalized analysis in view of the domestic and foreign related data integrati- on system research present situation, uses the global as view mapping and the local as view mapping separately, the integrity constraints and the query ques- tion has conducted the research according to data integration system based on relational model.Firstly, introduced the related concepts about data integration system, the main logarithm has conducted the thorough research according to the mapping mechanism as well as the integrity constraints in the data integration system, and then proposed a query preprocessing algorithm. The algorithm process the data integration system under integrity constraints based on the relational model which uses the local as view mapping, user query mapped on the new source views, consequently expanded the scope of query, and received more information.Secondly, according to the idea of data integration system under integrity constraints as well as constructed the canonical database, this paper proposed an algorithm whether the data integration system have the "restraint circula- tion". The algorithm process the foreign key constraints on the global schmal of the data integration system based on the relational model which use the global as view mapping. It can surmise whether the foreign key constraints have "restraint circulation" on data integration system. When the foreign key constraints in the data integration system does not have the "restraint circulat-ion", we can construct the limited canonical database. Not only solved the query question in data integration system under integrity constraints moreover enhanced the user query efficiency.