隐私信息泄露属性深度跟踪方法仿真

Simulation of Attribute Depth Tracking Method for Privacy Information Disclosure

由于网络规模普遍较大、用户数量较多,第三方域名采集用户信息时易造成隐私泄露问题,严重影响网络用户体验和满意度。提出一种基于动态密钥的深度跟踪方法。根据网络中社交群体内部关联性强、外部关联性弱的结构特性及公开用户隐私泄露风险大、非公开用户隐私泄露风险小特点,计算社区内部节点和外部节点间隐私关联,推测得到与隐私风险关联性较强的用户特征值。建立由密钥算子组成的信息泄漏跟踪模型,计算非公开用户的密钥参数,将其作为模型算子,提出隐私对照表,统计历史泄漏数据的属性值,与对照表数据对比输出判定结果,求解查找结果一致用户的位置信息,并实行组合排列,完成隐私泄露的深度跟踪。仿真结果表明,所提方法跟踪精准度高,在存在干扰和不存在干扰环境下均能达到预期效果,且该方法的运算量小,实用性更强。

Due to the generally large scale of the network and large number of users, third-party domain names are likely to cause privacy leakage when collecting user information, which seriously affects the user experience and satisfaction of the network. This paper presented a depth tracking method based on dynamic key. According to the structural characteristics such as strong internal correlation and weak external correlation of social groups in the network as well as high risk of privacy leakage of public users and low risk of privacy leakage of non-public users, the privacy association between internal nodes and external nodes of the community was calculated, and then users’ feature values with strong correlation of privacy risk were inferred. Moreover, an information leakage tracking model composed of key operators was built. And the key parameters of non-public users were calculated as the model operators. Furthermore, we designed a comparative table containing privacy for counting the attribute values of historical leakage data and compared them with the data in the comparative table, and output the judgment results. In the meanwhile, we solved the location information of users whose search results were consistent and implemented combination arrangement. Finally, we completed the depth tracking of privacy leakage. Simulation results show that the proposed method has high tracking accuracy, and can achieve the expected results no matter whether the interference exists or not. In addition, this method has less computation and stronger practicability.