Privacy-preserving data mining in peer to peer networks

In recent years, privacy-preserving data mining has been studied extensively, due to the wide increase of sensitive information on the internet. A number of algorithms and procedures have been designed, some of which are yet to be implemented, but a few of them are actually employed in the form of software systems to preserve the privacy of users, and the content in peer-to-peer networks. Privacy issues are becoming widely recognized when using peer-to-peer networks. In this paper, we provide a review of the privacy-preserving data mining techniques used in order to overcome privacy issues. We discuss methods of sanitization, data distortion, data hiding, cryptography and the data mining algorithm KDEC. Further discussion involves data transfer using proxy techniques, creating social communities among peer-to-peer users forming trusted peers. These techniques have shown to administer the issue of preserving data however show lack of scalability and performance. We design a framework to perform a comparison study on the techniques shown above and present the results with some recommendations of how we think the issues could be unraveled.