Research On Algorithm Of Privacy Preserving Data Mining

Data mining is the process of extracting cryptic, potential and helpful information from a mass of data. It is a kind of new technology for data analysis. At the same time, some confidential information will be leaked in the data mining process.As the enhancement of privacy protection consciousness,the problem of privacy protection has been paying more and more attention.As data mining technique has been becoming more and more important, how to protect privacy while mining data has become a hot research.This main job of this paper is to put forward two privacy protection data mining algorithms: one is the Randomized Response Decision Tree (RRDT) algorithm and the other is the distributed clustering algorithm based on short clusters.In the first chapter and the second chapter of the paper,we mainly introduce the background of our study and its present situation,also including some basic knowledge and algorithm.In the third chapter of this paper, we use the RRDT algorithm to disrupt the original data, then mining the data processed by the RRDT algorithm with association rules for privacy preserving.Theoretical analysis and the experiment shows that, the method based on RRDT privacy protection association rule mining is a very good one for the protection of data privacy, and also has a good performance.In the fourth chapter of this paper, we study the problem of secure multi-party computation. The aim is to protect the privacy data of participants and at the same time cooperate with each other well. At last, we dig out the desired results in the database of integration. We put forward a new algorithm based on privacy preserving by taking short clusters this concept into the environment of distributed secure multi-party. With theoretical analysis and comparing with the centralized K-means algorithm, we proved that this algorithm has a good performance of accuracy and safety.