隐私保护频繁项集挖掘中的细粒度随机化模型  

Fine-grained Randomized Response Model in Privacy Preserving Frequent Itemset Mining

作　　者：郭宇红[1] 童云海 GUO Yuhong;TONG Yunhai(School of Information Science and Technology,University of International Relations,Beijing 100091,China;Department of Intelligence Science,Peking University,Beijing 100871,China)

机构地区：[1]国际关系学院信息科技学院,北京100091 [2]北京大学智能科学系,北京100871

出　　处：《软件工程》2019年第10期44-46,43,共4页Software Engineering

基　　金：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)

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。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.

关 键 词：随机化回答 隐私保护 频繁项集 敏感问题调查 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]