Privacy Preserving Based on Vector Similarity for Weighted Social Networks
LAN Li-hui1,2, JU Shi-guang1
1. School of Computer Science and Telecommunication Engineering, Jiangsu University, Zhenjiang, Jiangsu 212013, China;
2. School of Information Engineering, Shenyang University, Shenyang, Liaoning 110000, China
Aiming at the publication of weighted social networks, a random perturbation method based on vector similarity is proposed.It can protect network structures and edge weights in multiple release scenarios.It constructs vector set models by segmentation based on vertex cluster using edge space theory.It adopts weighted Euclidean distance as similarity metrics to construct the released candidate sets according to the threshold.It randomly selects vectors from candidate sets to construct the published weighted social networks.The proposed method can resist multiple vertex recognition attacks, force attackers to re-identify in a large result set that the existential probabilities of the vectors are same, and increase the uncertainty of recognition.The experimental results demonstrate that it can preserve individuals' privacy security, meanwhile it can protect some structure characteristics for networks analysis and improve data utility.
兰丽辉, 鞠时光. 基于向量相似的权重社会网络隐私保护[J]. 电子学报, 2015, 43(8): 1568-1574.
LAN Li-hui, JU Shi-guang. Privacy Preserving Based on Vector Similarity for Weighted Social Networks. Chinese Journal of Electronics, 2015, 43(8): 1568-1574.
[1] Zhou B,Pei J.Preserving privacy in social networks against neighborhood attacks[A].2008 IEEE 24th International Conference on Data Engineering[C].Cancun,Mexico:IEEE Computer Society,2008.506-515.
[2] Liu K,Terzi E.Towards identity anonymization on graphs[A].2008 ACM SIGMOD International Conference on Management of Data[C].Vancouver,Canada:Association for Computing Machinery,2008.93-106.
[3] Zou L,Chen L,Özsu M T.K-automorphism:A general framework for privacy preserving network publication[J].Proceedings of the VLDB Endowment,2009,2(1):946-957.
[4] Hay M,Mikiau G,Jensen D,Weis P,et al.Anonymizing social networks[R].University of Massachusetts Amherst,2007.
[5] Ying X W,Wu X T.Randomizing social networks:A spectrum preserving approach[A].2008 8th SIAM International Conference on Data Mining[C].Atlanta,United States:Society for Industrial and Applied Mathematics Publications,2008.739-750.
[6] Liu L,Wang J,Liu J et al.Privacy preserving in social networks against sensitive edge disclosure[R].Department of Computer Science,University of Kentucky,2008.
[7] Das S,Egecioglu Ö,Abbadi A E.Anónimos:An LP-based approach for anonymizing edge-weighted social network graphs[J].IEEE Transactions on Knowledge and Data Engineering,2012,4(4):590-603.
[8] Li Y,Shen H.Anonymizinggraphs against weight-based attacks[A].2010 10th IEEE International Conference on Data Mining Workshops[C].Sydney,Australia:IEEE,2010.491-498.
[9] Skarkala M E,Maragoudakis M,Gritzalis S et al.Privacypreservation by k-anonymization of weighted social networks[A].2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining[C].Istanbul,Turkey:IEEE Computer Society,2012.423-428.
[10] 王树禾.图论(第二版)[M].北京:科学出版社,2009:177-180.
[11] 吕锋,杜妮,文成林.一种模糊-证据kNN分类方法[J].电子学报,2012,40(12):2390-2395. Lü Feng,Du Ni,Wen Cheng-lin.A fuzzy-evidential k nearest neighbor classification algorithm[J].Acta Electronica Sinica,2012,40(12):2390-2395.( in Chinese)
[12] Newman M E J,Girvan M.Finding and evaluating community structure in networks[J].Physical Review E,2004,69(2):1-15.
[13] Lü L,Zhou T.Link prediction in complex networks:A survey[J].Physica A:Statistical Mechanics and its Applications,2011,390(6):1150-1170.
[14] Girvan M,Newman M E J.Community structure in social and biological networks[J].Proceedings of the National Academy of Sciences,2002,99(12):7821-7826.
[15] Nexus.The network repository [DB/OL].http://nexus.igraph.org/.
2015, Vol. 43 
