论文部分内容阅读
针对垂直划分多决策表,利用半可信第三方和交换加密体制,设计了一个安全多方计算交集基数协议。利用该协议设计了安全多方计算信息熵和安全多方计算条件信息熵的解决方案,提出了一种基于条件信息熵的隐私保护属性约简算法。该算法基于粗糙集信息观的约简理论实现了分布式环境下全局属性约简的求解,使各参与方在不共享其隐私信息的前提下达到集中式属性约简的效果。分析结果表明该算法是有效可行的。
Aiming at the multi-decision-making table of vertical partitioning, a secure multi-party computing intersection base protocol is designed by using semi-trusted third party and exchange encryption system. This protocol is used to design a solution to secure multi-party computing information entropy and secure multi-party computing conditional information entropy. A privacy preserving attribute reduction algorithm based on conditional information entropy is proposed. The algorithm realizes the solution of global attribute reduction in distributed environment based on the reduction theory of rough set information view, so that each participant can achieve the effect of centralized attribute reduction without sharing its private information. The analysis results show that the algorithm is effective and feasible.