Published Paper Details:

Many data owners are required to release the data in a variety of real world application, since it is of vital importance to discovery valuable information stay behind the data. In existing re-identification attacks on the health care data sets. Such data publish directly to the world that may violate the individual privacy. It is urgent to solve the re-identification risks by using efficient recommendation of de-identification policies to guarantee both privacy and utility of the sensitive data. Using De-identification policies to achieve such requirements, at the same time number of de-identification policies is large due to the quasi-identifier attributes. To control the tradeoff between data utility and data privacy, to select policies skyline computation techniques is used but it is challenging for large number of policies. In this paper, we propose algorithm known Sky-Filter-MR. It is fully based on Map Reduce technique. De-identification policies are used to overcome the challenge of the skyline computing in large policies that are represented by in the form bit-strings. A novel skyline computation scheme was used to remove unqualified policies using the approximate relationship for further improve the performance. In our proposed SKY-FILTER-MR algorithm are Extensive experiments in both real life datasets are out perform approach by up to four times faster. It indicates good scalability in large policy datasets.