Collusion-Resistant Privacy-Preserving Data Mining

抄録

Numerous methods have been proposed for privacy-preserving data mining (PPDM). Most of these methods are based on an assumption that semi-honest is and collusion is not present. In other words, some parties may collude and share their record to deduce the private information of other parties. We considered a general problem in this issue - multiparty secure computation of functions on secure summations of data spreading around multiple parties. In order to solve above collustion problem, we have proposed a secure computation method that entails a high level of collusion-resistance. Unfortunately, the private inputs of some parties may be infered because unnecessary information is disclosed in the process of this method. In this paper, we will improve this method, so that the final result is directly computed without any intermediate information being revealed. Moreover, this method can be used to securely compute almost any kind of function on secure cummations.