基于聚类相关性约束的（s,l）-多样性匿名方法

摘要

In allusion to the problem of traditional data anonymity models constructing equivalence class with high correlative sensitive values, (s,l)-diversity was proposed which limited the correlativity of sensitive values in the equivalence classes. This diversity model was based on traditionall-diversity model, and it measured the correlativity of the sensitive attribute values to decrease the information loss by equivalence classes with high corrective sensitive values. At the same time, a (s,l)-diversity clustering algorithm named SLCA was proposed to achieve (s,l)-diversity, and the SLCA algorithm measured the distance between tuples by measuring the correlativity of attribute values, which greatly decreased the information loss during data generation. The results show that SLCA algorithm is more effective in terms of both information loss and execution time, and SLCA algorithm can effectively decrease the correlativity of the sensitive values in the equivalence classes to protect the privacy security of the data sets.%针对传统l-多样性模型易形成敏感值高度相关的等价类问题，提出一种约束等价类中敏感值相关性的(s,l)-多样性模型。该模型在传统l-多样性模型的基础上，以敏感集合中非敏感属性值的分布度量敏感值的相关性，通过等价类中敏感值相关性的约束来降低高相关性敏感值产生的信息泄露。同时，使用属性值间相关性作为距离度量基准，提出一种(s,l)-多样性聚类算法(SLCA)来实现该匿名模型，以降低数据泛化过程中的信息损失。研究结果表明：SLCA算法具有较小的信息损失量与较短的运行时间，能够有效地降低等价类中敏感值的相关性，更好地防止个体敏感信息泄露。