On the complexity and approximability of repair position selection problem

Inconsistent data indicates that there is conflicted information in the data, which can be formalized as the violations of given semantic constraints. To improve data quality, repair means to make the data consistent by modifying the original data. Using the feedbacks of users to direct the repair operations is a popular solution. Under the setting of big data, it is unrealistic to let users give their feedbacks on the whole data set. In this paper, the repair position selection problem (RPS for short) is formally defined and studied. Intuitively, the RPS problem tries to find an optimal set of repair positions under the limitation of repairing cost such that we can obtain consistent data as many as possible. First, the RPS problem is formalized. Then, by considering three different repair strategies, the complexities and approximabilities of the corresponding RPS problems are studied.