在线社交网络中模板克隆的IAC检测方法
An IAC Approach for Detecting Profile Cloning in Online Social Networks
摘要：
如今，在线社交网络是互联网上流行的网站，数以百万计的用户在其上注册并与他人分享他们自己的个人信息。隐私威胁和泄露个人信息是OSN用户最关心的问题。近年来，在OSN上发现了一种新的攻击方式，即身份克隆攻击。在这种攻击中，攻击者试图伪造真实用户的身份，以便访问用户朋友的私人信息，而这些信息并没有发布在公共配置文件上。在今天的OSN中，有一些验证服务，但它们不是主动服务，它们对熟悉在线身份问题的用户很有用。本文对身份克隆攻击进行了详细的分析，并提出了一种新的、精确的在线社交网络中身份克隆检测方法。该方法首先将社会网络以图的形式表示出来，然后根据用户之间的相似性将图划分为更小的社区。然后，收集所有与真实简档相似的简档（来自同一社区），然后计算关系强度（在所有选择的简档和真实简档之间），并且那些具有较小关系强度的简档将通过共同好友系统进行验证。在这项研究中，为了评估所提出的方法的有效性，所有步骤都应用在Facebook的数据集上，最后通过将它们应用到数据集上，将这项工作与以前的两项工作进行比较。
英文摘要：
Nowadays, Online Social Networks are popular websites on the internet, which millions of users register on and share their own personal information with others. Privacy threats and disclosing personal information are the most important concerns of OSNs users. Recently, a new attack which is named Identity Cloned Attack is detected on OSNs. In this attack the attacker tries to make a fake identity of a real user in order to access to private information of the users friends which they do not publish on the public profiles. In today OSNs, there are some verification services, but they are not active services and they are useful for users who are familiar with online identity issues. In this paper, Identity cloned attacks are explained in more details and a new and precise method to detect profile cloning in online social networks is proposed. In this method, first, the social network is shown in a form of graph, then, according to similarities among users, this graph is divided into smaller communities. Afterwards, all of the similar profiles to the real profile are gathered (from the same community), then strength of relationship (among all selected profiles and the real profile) is calculated, and those which have the less strength of relationship will be verified by mutual friend system. In this study, in order to evaluate the effectiveness of proposed method, all steps are applied on a dataset of Facebook, and finally this work is compared with two previous works by applying them on the dataset.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
