A Novel Algorithm for Identifying Corresponding Attributes in Heterogeneous Databases

Identifying corresponding attributes is an important issue to realize data sharing and interoperability in heterogeneous databases. The main approaches at present adopted predefined rules to evaluate the similarity of attributes by comparing all attributes, which have some limitations. So two-phase-checking algorithm based on BP neural network was presented in this paper to realize attributes matching. The experimental results show our proposed approach can improve the system performance, the precision and recall of attributes matching obviously.