Identifying Corresponding Entities Based on Attribute Entropy in Heterogeneous Databases

With progress of information technology, it is necessary to access multiple heterogeneous databases in order to get the desired information from different sources. One main problem encountered constantly among heterogeneous databases is entity heterogeneity, which arises when the same real-word entity type is represented using different identifiers in different application domains. So identifying the heterogeneous entities and resolving conflicts are critical when realizing interoperability in multiple heterogeneous databases. In this paper, we examine the entity heterogeneity problem and propose an approach based on attribute entropy to identifying corresponding entities. The experimental results on real-world data show our proposed approach is very effective.