| 摘 要: 已有的随机化回答模型调控的数据范围宽、粒度粗，对隐私数据的保护粒度缺乏灵活性，无法实现精细 化、个性化、差异化的隐私保护。提出三类多参数随机化回答模型，包括行多参、复合多参、分组多参共11种随机化回 答模型，给出了模型的分类框架和分类层次。细粒度多参数随机化模型可实现精细化、个性化、差异化的隐私保护 效果。 |
| 关键词: 随机化回答；隐私保护；频繁项集；敏感问题调查 |
|
中图分类号: TP311
文献标识码: A
|
| 基金项目: 论文由国际关系学院中央高校基本科研业务费项目资助(项目编号3262017T48,3262018T02,3262019T06). |
|
| Fine-grained Randomized Response Model in Privacy Preserving Frequent Itemset Mining |
|
GUO Yuhong,TONG Yunhai1,2
|
1.( 1.School of Information Science and Technology, University of International Relations, Beijing 100091, China;2.
2.Department of Intelligence Science, Peking University, Beijing 100871, China)
|
| Abstract: The existing randomized response model regulates a wide range of data with coarse granularity and lacks flexibility in protecting privacy,unable to achieve fine,personalized and differentiated privacy protection.Three kinds,11 types of multi-parameter random response models are proposed,including row multi-parameter,compound multi-parameter and grouping multi-parameter.The classification framework and hierarchy of these models are given.The fine-grained multiparameter randomized response models can realize fine,individualized and differentiated privacy protection effect. |
| Keywords: randomized response;privacy preserving;frequent item set;sensitivity survey |
