The rise of mobile networks as well as intelligent terminal has generated a lot of data ,studying and analyzing such data is essential for scientific researcher and business development .However ,it will be a serious threat to the user's privacy ,especially the sensitive information ,if adversary is able to identify and steal user's privacy .The existing privacy protection methods usually consider sensitive attributes and corresponding values from an overall perspective ,but the sensi‐tivity of different sensitive attributes and different requirements for user's specific sensitive attributes are paid less attention to .In this paper ,an improved algorithm based on K‐anonymity model is proposed .In addition ,based on information entro‐py ,the degree of anonymity of this algorithm is evaluated .%移动网络的兴以及智能终端产生了大量的数据，研究与分析这样的数据有利于科学研究、商业发展等。然而，如果攻击者能够识别或窃取数据中的用户隐私信息，特别是敏感信息，给用户个人隐私信息带了严重的威胁。现有隐私保护技术从整体角度考虑敏感属性及取值，却较少考虑到不同用户对某个敏感属性的敏感程度以及相应的隐私保护需求的综合影响。论文提出一种基于 K‐匿名模型的改进算法，同时考虑不同敏感属性的整体敏感度以及用户对具体敏感属性的不同需求，实现基于多敏感值的个性化隐私保护算法。最后通过信息熵评价算法的匿名程度。
展开▼
