摘要
针对社交网络信息推荐中的信息传播带来隐私泄露的问题,结合信息传播模型,提出了一种支持隐私保护的社交网络信息推荐方法,通过好友的兴趣度、熟悉度和兴趣相似度推测用户兴趣,进行文本匹配和推荐候选集排序;通过个性化隐私偏好设置允许用户设置受限访问用户列表,并使用隐私保护方法计算信息传播至黑名单用户的概率,设置隐私泄露阈值对黑名单用户访问隐私博文的概率进行控制,达到信息推荐中保护用户隐私的目的。实验结果表明,所提方法可以在保证推荐效果的同时更好地保护用户隐私。
Aiming at the problem of privacy disclosure caused by information dissemination in social network information recommendation, a kind of social network information recommendation method that supports privacy protection is proposed in combination with information dissemination model. Users′ interests are predicted by friends′ interest, familiarity and interest similarity, and then text matching and recommendation candidate set sorting are carried out. Through personalized privacy preference setting, users are allowed to set restricted access to user list, and privacy protection method is used to calculate the probability of information spreading to blacklist users, and privacy disclosure threshold is set to control the probability of blacklist users accessing privacy blog posts to protect user privacy in information recommendation. Experimental results show that the proposed method can guarantee the recommendation effect and protect user privacy better.
作者
张超
梁英
方浩汕
ZHANG Chao;LIANG Ying;FANG Hao-shan(Research Center for Ubiquitous Computing Systems,Institute of Computing Technology,Chinese Academy of Sciences,Beijing 100190,China;School of Artificial Intelligence,University of Chinese Academy of Sciences,Beijing 100049,China;School of Software Engineering,Shandong University,Jinan 250101,Shandong,China)
出处
《山东大学学报(理学版)》
CAS
CSCD
北大核心
2020年第3期9-18,共10页
Journal of Shandong University(Natural Science)
基金
国家重点研发计划项目(2018YFB1004704,2016YFB0800403)。
关键词
社交网络
信息推荐
隐私保护
访问控制
数据挖掘
social network
information recommendation
privacy protection
access control
data mining