Privacy Preserving Clustering by Random Response Method of Geometric Transformation
Jie Liu, Yifeng Xu · 2009
With the large influx of the data mining technology and mining tools, the confidentiality requirements of the personal privacy are becoming more and more urgent. Therefore, how to ensure personal privacy and get the correct mining results becomes a severe issue to be resolved. In this paper, we propose a kind of random response method of geometric transformation- the combination of the random response technology and the geometric transform algorithm. The algorithm is designed to solve the shortage of low privacy protection of the geometric transform algorithm. The algorithm first gives four parameters, corresponding to the probability of four different types of geometric transformations. According to the various random number generated, different geometric transformation method is selected, which serves the dual effect of privacy protection. Our experiment proves that this method has a high degree of privacy protection and can get correct mining results.