A Flexible Method of Privacy Preserving Clustering

Privacy protection is an important issue in data processing. In this paper, we present a novel clustering method for privacy preserving in homogenous data sets. By developing matrix transformation method, our method can not only protect privacy in face of collusion, but also achieves a higher level of accuracy as compared to the existing method. The importance of independent perturbation is addressed in the random matrix generation. The performance of the method including the levels of accuracy and privacy are also analyzed in detail. Experimental results further demonstrate that our method is also adaptive to large data dimensions.