Data disturbance and reconstruction system, data reconstruction apparatus, data reconstruction method, data reconstruction program

摘要

PROBLEM TO BE SOLVED: To reduce processing cost for data restructuring.SOLUTION: The data disturbance and restructuring system of the present invention comprises a data disturbance apparatus and a data restructuring apparatus. The data disturbance apparatus generates a cross tabulation Y for a disturbance table from an initial table composed of N records having K attributes by using a transition probability matrix A that disturbs attribute values of only a part of the attributes. The data restructuring apparatus comprises a value range calculation unit, a matrix generation unit, a vector generation unit and an iterative Bayesian unit. The matrix generation unit generates, for each combination of the attribute values of holding attributes, a Q×Q partial transition probability matrix Afrom a component of the transition probability matrix A. The vector generation unit generates a Q-order vector Yfrom a component of the cross tabulation Y. The iterative Bayesian unit obtains a Q-order vector Xthat indicates a restructured cross tabulation for a disturbed attribute, from the partial transition probability matrix Aand the vector Y. A cross tabulation X is restructured using all vectors X.