摘要: 针对分组无关问题模型存在隐私泄露的问题,提出一种改进的分组无关问题模型,采用随机响应的方法,通过对原始数据进行伪装变换处理,实现具有隐私保护的关联规则挖掘。实验结果表明,改进后的模型在伪装变换后的数据集上挖掘出的规则与原始数据规则相比,保证了低误差,具有较好的隐私保护性。
关键词:
分组无关问题模型,
随机响应,
关联规则挖掘
Abstract: Aiming at the problem of privacy leak exists in the grouping unrelated-question model, this paper presents an improved grouping unrelated-question model to achieve the Privacy-preserving association rules through disguising and changing the original data. Experimental results show that the rule which algorithm gets has a lower error and better privacy compared with the original rule.
Key words:
grouping unrelated-question model,
random response,
association rules mining
中图分类号:
阚莹莹;曹天杰. 基于分组无关问题模型的隐私保护算法[J]. 计算机工程, 2010, 36(5): 79-80,8.
KAN Ying-ying; CAO Tian-jie. Privacy-preserving Algorithm Based on Grouping Unrelated-question Model[J]. Computer Engineering, 2010, 36(5): 79-80,8.