中文摘要：

针对聚类分析时如何保护隐私的问题，从传统的数据安全度评价标准出发，重新拓展了一般实数上有限维欧氏空间中隐私保护度的评价指标，提出了一种称为OBT（基于正交变换的数据转换方法）的算法，OBT中正交矩阵的选择不依赖于具体数据，能够很好地应用于大容量的数据库上，在应用正交变换保护数据中的隐私信息时不需要进行大量的运算。

英文摘要：

This paper is concentrated on the issue of protecting the underlying attribute values when sharing data for clustering and systemically discuss the process which preserves privacy information by Orthogonal Transformation. On the basis of traditionally measure of security, this paper extends the measure of privacy preserving in Euclidean space and proposes a privacy preserving method called Orthogonal-Based Transformation （OBT）. OBT doesn＇ t depend on concrete data and it makes the process neednt a great deal computation which preserves privacy information by Orthogonal Transformation.