Privacy preserving clustering over distributed data
Fanrong Meng, Bin Liu, Chujiao Wang · 2010
Data mining based on privacy preserving is the combination of information security technology and knowledge discovery technology. A simple and effective privacy-preserving distributed mining method of clustering (PPD-SMD) and (PPD-JD) is proposed to solve the issue about privacy preserving of cluster based on Binary and Nominal Attributes distance. This method brings the secure protocol and crypto-algorithm in the data models of the horizontal distributed. Using semi-trusted third party (STTP), PPD-SMD and PPD-JD do not transfer real data to other sites in clustering procedure. In the end, analysis in security, efficient and complexity are carried on.