,Embedding differential privacy in decision tree algorithm with different depths

来源 :中国科学(信息科学) | 被引量 : 0次 | 上传用户:algo12
下载到本地 , 更方便阅读
声明 : 本文档内容版权归属内容提供方 , 如果您对本文有版权争议 , 可与客服联系进行内容授权或下架
论文部分内容阅读
Differential privacy (DP) has become one of the most important solutions for privacy protection in recent years.Previous studies have shown that prediction accuracy usually increases as more data mining (DM) logic is considered in the DP implementation.However,although one-step DM computation for decision tree (DT) model has been investigated,existing research has not studied the scenarios when the DP is embedded in two-step DM computation,three-step DM computation until the whole model DM computation.It is very challenging to embed DP in more than two steps of DM computation since the solution space exponentially increases with the increase of computational complexity.In this work,we propose algorithms by making use of Markov Chain Monte Carlo (MCMC) method,which can efficiently search a computationally infeasible space to embed DP into DT generation algorithm.We compare the performance when embedding DP in DT with different depths,i.e.,one-step DM computation (previous work),two-step,three-step and the whole model.We find that the deep combination of DP and DT does help to increase the prediction accuracy.However,when the privacy budget is very large (e.g.,e =10),this may overwhelm the complexity of DT model,and the increasing trend is not obvious.We also find that the prediction accuracy decreases with the increase of model complexity.
其他文献
正确使用壮丰灵是当前玉米田增产增收的最佳途径赵静峰,李淑娥,张新忠(梨树县梨树乡农业站)自丹育13后的几年来,我县先后从省内外引进了许多替代品种,但都因其贪青、倒伏等原因造成产
This paper focuses on the distributed filtering problem for a class of time-varying networked systems with sensor gain degradation and energy constrained commun
Human induced pluripotent stem cells (iPSCs) hold unprecedented potential to model human development and genetic disorders due to their capability to self-renew
文章针对我国室内环境艺术设计发展历程,通过分析教育与社会需求的对应关系,以及中国室内设计领域现存问题的根源,提出艺术教育的本源和学科总体发展的定位,试图探求适应于本
近年来,由于人类活动的影响,氮沉降等全球变化问题逐年加剧,导致氮沉降对土壤有机碳库产生影响。相对于氮沉降量的影响,我们对于不同形态氮沉降对土壤有机碳如何影响及其影响的生
With the popularity of location based service (LBS),a vast number of trust medels for LBS recommendation (LBSR) have been proposed.These trust models are centra
Precision medicine emerges as a new approach that takes into account individual variability.Successful realization of precision medicine requires disease models
北京市1989年高等教育自学考试新闻理论课程的考试和试卷评阅工作已告结束。从卷面看,总的情况好于去年,及格率比去年提高9%。不少考生在知识面、基本理论、基本技能技巧等
三峡库区是我国水土流失最为严重的地区之一,库区土壤侵蚀产生的泥沙淤积和面源污染是水库安全运行和效益发挥的主要环境问题。为了防治库区水土流失,急需开展土壤侵蚀定量评价
学位