期刊文献+

跨社交平台的用户识别方法研究

Research and Implementation of User Linkage Technology across Social Networks
下载PDF
导出
摘要 随着互联网及应用的快速发展,社交网络对人们的生活影响越来越大.一个用户可能同时具有多个社交网络账号.如果能够关联出同一用户在不同社交网络中的账号,对于网络安全监管有很大帮助.对主流社交平台Facebook和Twitter采用基于账号属性信息结合熵权法的方法进行用户身份关联.通过分析以及融合用户数据,选择出用户名、个人主页、地理位置和个人描述四个属性作为用户身份关联的特征,采用不同的相似度匹配算法计算四个属性的相似度,并利用熵权法计算出最终的相似度,最后根据决策判定方法确定出关联结果.实验结果表明,基于属性信息用户身份关联方法具有一定的用户身份关联能力,关联结果召回率可以达到90.26%,为社交网络的用户身份关联提供参考. With the rapid development of the Internet and Application,social networks affect people’slife a lot,and an user maybe own more than one social networks accounts in the meantime.If it is possible to identify the same user with different social network accounts,it would be bene ficial to network security supervision.Selects two mainstream social networks,Facebook and Twitter,to conduct experiments based on the attributes and entropy method.After analysis and fusion of user data,four characteristic of user name,personal homepage,geographic loca tion,and personal description are selected as attributes of user identification.Different similarity matching algorithms are respectively used to calculate the similarity of four attributes and different entropy weights are given to the corresponding attributes.And the total similarity between users is obtained based on the four attributes and their weights.Finally,the recognition result is determined according to the deci sion-making method,that is,the same user's account in different social networks.The experimental results show that the user identification method adopted in this paper has a certain ability of user identification,and the accuracy rate of the recognition result can reach 90.26%,of fering references for identifying users on multiple social networks.
作者 文玥琪 周安民 WEN Yue-qi;ZHOU An-min(College of Cyberspace Security,Sichuan University,Chengdu 610225)
出处 《现代计算机》 2020年第8期37-42,共6页 Modern Computer
关键词 社交网络 用户身份关联 属性相似度 熵权法 Social Network User Identity Linkage Attribute Similarity Entropy Method
  • 相关文献

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部