Algorithm for k-anonymity based on projection area density partition

Abstract

In data publishing privacy preserving,while classifying temporary anonymous groups,the existing algorithms didn’t consider the distance between adjacent data points,and could easily produce a lot of unnecessary information loss,thus affecting the availability of released anonymous data sets.To solve the above problem,the concept of rectangular projection area,the projection area density and partition coefficient characterization were presented,aim to increase the recording points’s projection area density to divide temporary anonymous group reasonably,and to make the information loss of divided anonymous groups as small as possible.And presents the algorithm for k-anonymity based on projection area density partition,by optimizing the rounded partition function and properties dimension selection strategy,to reduce unnecessary information loss and to further improve the availability of released data sets,without reducing the number of anonymous groups.The rationality and validity of the algorithm are verified by theoretical analysis and multiple experiments.

Keywords

• privacy preserving;temporary anonymous group;rectangular projection area;projection area density