文献类型：期刊文献

英文题名：Published weighted social networks privacy preservation based on community division

作者：Zhang, Xueqin[1]; Zhou, Qianru[1]; Gu, Chunhua[1]

机构：[1] School of Information Science and Engineering, East China University of Science and Technology, Shanghai, 200237, China

年份：2017

起止页码：86

外文期刊名：ACM International Conference Proceeding Series

收录：EI(收录号：20180704798544)

语种：英文

外文关键词：Social networking (online) - Data privacy - Publishing - Undirected graphs

摘要：Social network not only contains privacy information but also contains large valuable data for researching. It is very critical to balance the utility of data and the power of privacy preservation for published network. This paper proposes privacy preservation based on community division algorithm to preserve the privacy of published weighted social network. The algorithm firstly changes the directed graph into undirected graph whose weight is replaced by relationship strength and takes the node similarity into consideration while dividing communities, then perturb randomly in local structures to generate the published graph. NMI is used to verify the accuracy of community division, the results of three datasets prove that the algorithm can protect the privacy and guarantee the data utility of the graph structure at the same time. ? 2017 Association for Computing Machinery.