Framework for Privacy-Preserving Data Mining

Project Discription

The introduction of the decisions of discriminatory was first time given by Pedreschi. The method is dependent on rules of mining classification i.e. the part of inductive and their reasoning i.e. the deductive-part is dependent on the significant measures of discrimination which characterize the genuine definitions of the discrimination. For this purpose, the US Equal Pay Act states that: “a selection rate for any race, gender, or specific group which is less than four-fifths of the rate for the group with the highest rate will generally be regarded as evidence of adverse impact.” This mechanism has been upgrade to enclose the statistical importance of derived patterns of the discrimination within paper and to the causes regarding the affirmative activity and favoritism.

Additionally as a tool dependent on Oracle that has been applied in the paper. Recent discovery methods for discrimination take every rule separately for evaluating the discrimination without any consideration to another rules or association in between them. However, work is also taken as the association between the rules for discovery of the discrimination which is dependent on the nonexistence or existence of the discriminatory elements.

Discrimination-prevention, it is the another main anti-discrimination objective in the data-mining, that includes the inducing patterns which do not directs to the decisions of discriminatory as if original sample data-sets are partially included.

Project Sample Image