Information Mining Research Paper

IV. Information Miner

A. CONCERNS OF INFORMATION MINER

With a sorts of data hidden the information. Some of the time the information mining results may uncover delicate data about the information proprietors. For instance, in the Target story we said in Section I-B, the data about the girl's pregnancy, which is derived by the retailer by means specific end goal to find valuable learning which is fancied by the leader, the information digger applies information mining calculations to the information got from information authority. The security issues accompanying the information mining operations are twofold. On one hand, if individual data can be straightforwardly seen in the information and information rupture happens, protection of the first information proprietor (i.e. the information supplier) will be bargained. Then again, outfitting with the numerous effective information mining procedures, the information excavator can find out different of mining client information, is something that the little girl does not need others to know. To

Reconstruction-based association rule hiding.

These approaches can roughly be categorized into the following five groups:

Heuristic distortion approaches, which resolve how to select the appropriate data sets for data modification.

Heuristic blocking approaches, which reduce the degree of support and confidence of the sensitive association rules by replacing certain attributes of some data items with a specific symbol (e.g. `?').

Probabilistic distortion approaches, which distort the data through random numbers generated from a predefined probability distribution function.

Exact database distortion approaches, which formulate the solution of the hiding problem as a constraint satisfaction problem (CSP), and apply linear programming approaches to its solution.

Reconstruction-based approaches, which generate a database from the scratch that is compatible with a given set of non-sensitive association