Probabilistic Query Rewriting in Inconsistent Databases

This paper combines with original probabilistic databases based on the clusters which is produced by tuple matching techniques,and proposes a new probabilistic approach based on clustering in inconsistent databases.It analyzes a basic technique for queries rewriting based on believable cluster,considers the appropriate tuple probabilities,interval values and expectation values.The approach needs not procedural pre-processing,so the rewritten queries can be efficiently optimized and executed by commercial database systems.In order to understand the flexibility of the approach for considering distinguishing degrees of inconsistency and database sizes,the experiments use the data and queries of the TPC-H specification,and results show that the approach is efficient.