[1]吉根林,姚? 瑶.一种分布式隐私保护的密度聚类算法[J].智能系统学报,2009,4(2):137-141.
JI Gen-lin,YAO Yao.Densitybased privacy preserving distributed clustering algorithm[J].CAAI Transactions on Intelligent Systems,2009,4(2):137-141.
点击复制
《智能系统学报》[ISSN 1673-4785/CN 23-1538/TP] 卷:
4
期数:
2009年第2期
页码:
137-141
栏目:
学术论文—机器学习
出版日期:
2009-04-25
- Title:
-
Densitybased privacy preserving distributed clustering algorithm
- 文章编号:
-
1673-4785(2009)02-0137-05
- 作者:
-
吉根林,姚? 瑶
-
南京师范大学数学与计算机科学学院,江苏南京210097
- Author(s):
-
JI Gen-lin, YAO Yao
-
School of Mathematics and Computer Science, Nanjing Normal University, Nanjing 210097, China
-
- 关键词:
-
隐私保护; 分布式聚类; DBDC; DBPPDC
- Keywords:
-
privacy preserving; distributed clustering; DBDC; DBPPDC
- 分类号:
-
TP311.1
- 文献标志码:
-
A
- 摘要:
-
对基于密度的分布式聚类算法DBDC进行改进,提出了一种基于密度的分布式隐私保护聚类算法DBPPDC.在由局部模型确定全局模型时,通过相关安全协议有效地保护了局部模型,同时不影响全局聚类.在利用全局模型更新局部模型时,通过改进算法、应用安全协议保护隐私信息,最终使各站点分布的数据能够安全聚类.理论分析和实验结果表明,DBPPDC算法是有效的.
- Abstract:
-
A densitybased privacy preserving distributed clustering algorithm (DBPPDC) was proposed following the improvements to the densitybased distributed clustering DBDC algorithm. When a global model is determined from a local model, (DBPPDC) effectively protects the local model without obstructing global clustering. On the contrary, when the local model is updated with the global model, DBPPDC makes all the data in local sites cluster safely by improving the previous algorithm and appling a secure protocol. Experimental results showed that DBPPDC is effective and efficient.
更新日期/Last Update:
2009-05-04