【摘要】 提出一种(θ,k)-匿名模型,通过对记录进行语义分析确定敏感属性值的相似或相异性,将一个确定了k值的等价类分成θ组,使记录在组内保持敏感属性值相似,在组间保持敏感属性值相异,并采用距离度量方法划分等价类.实验结果表明,(θ,k)-匿名模型可以在较低的信息损失下,同时抵制背景知识与相似性双重攻击.更多还原

【Abstract】 A( θ,k)-anonymous model to determine the similarity or dissimilarity of sensitive attribute values through semantic analysis of records was proposed. An equivalence class with determined k value was divided into θ groups,which made the records similar in terms of sensitive attribute values within the group,and kept sensitive attribute values varied among groups. Distance measurement method was used in partition. Experimental results showed that the( θ,k)-anonymous model could resist background knowledge attack and similarity attack at the same time with lower information loss.更多还原