一种基于k-means的分布式k-anonymity算法

With the arrival of big data era, there is increasing demand upon data sharing and data release. However, publishing or sharing raw data without anonymous processing would result in user privacy disclosure. Therefore, k-anonymity is widely used as a basic principle for above problems. In big data scenarios, single computing node is not able to meet the needs of k-anonymity computing, which makes the study of distributed k-anonymity algorithm particularly important. Based on k-anonymity principle and k-means algorithm, this paper proposed a binary k-clustering algorithm as distributed solution. Furthermore, this paper proposed quick binary k-clustering algorithm for time complexity optimization. The result of algorithm tests shows that the quick binary k-clustering algorithm has obvious time optimization as well as lesser degree of data loss compares to heuristic distributed k-anonymity algorithm.