基于paillier的隐私保护下关联规则挖掘方法

摘要

In privacy preserving association rule mining, the precision and security of mining are always a pair of contradictions. A method of privacy-preserving mining of association rules based on paillier encryption algorithm over distributed databases was proposed. The method separated calculation and decryption so it can solve the prob-lem of accuracy and security from mining of association rules perfectly. The method can reduce the time cost by Montgomery reduction. The experiment shows that the time cost on the basis of adding the process of encryption and decryption is acceptable.%在基于隐私保护的数据挖掘中，挖掘的精度和安全性一直是一对矛盾，针对该矛盾提出了一种在分布式的环境下基于paillier同态加密的关联规则挖掘方法，该方法将计算方和解密方分离保证了数据挖掘的安全性，同时基于同态加密并不会影响挖掘的精度，并通过蒙哥马利算法降低了paillier算法的开销，经过实验发现在增加加解密过程后算法开销还是在能接受的范畴内。