作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2006, Vol. 32 ›› Issue (13): 12-14. doi: 10.3969/j.issn.1000-3428.2006.13.005

• 博士论文 • 上一篇    下一篇

分布式环境下保持隐私的关联规则挖掘算法

黄毅群;卢正鼎;胡和平;李瑞轩   

  1. 华中科技大学计算机学院,武汉 430074

  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2006-07-05 发布日期:2006-07-05

Privacy Preserving Distributed Data Mining Association Rules of Frequent Itemsets

HUANG Yiqun;LU Zhengding;HU Heping; LI Ruixuan

  

  1. School of Computer Science & Technology, Huazhong University and Technology, Wuhan 430074)

  • Received:1900-01-01 Revised:1900-01-01 Online:2006-07-05 Published:2006-07-05

摘要:

保持隐私是未来数据挖掘领域的焦点问题之一,如何在不共享精确数据的条件下,获取准确的数据关系是保持隐私的数据挖掘的首要任务。该文介绍了分布式环境下保持隐私的数据挖掘的基本问题和措施,研究了一种基于向量点积的关联规则挖掘算法,给出了一种安全的向量点积协议。对于垂直划分的分布式数据库,该协议既可用于搜索频繁项集,又能保持各方数据的隐私。

关键词: 保持隐私, 分布式数据挖掘, 关联规则, 频繁项集, 点积

Abstract:

There has been growing interests in private concerns for future data mining research. Privacy preserving data mining concentrates on developing accurate models without sharing precise individual data records. This paper addresses basic ideas and solutions for secure data mining over distributed data. An algorithm based on dot product for distributed mining association rules is presented. It also gives a protocol of secure dot product computation which is effective to discover frequent itemsets on vertically partitioned data. It can provide good data privacy.

Key words: Privacy preserving, Distributed data mining, Association rules, Frequent itemsets, Dot product