网络安全中基于用户群体偏好的隐私保护算法

Privacy Protection Algorithm Based on User Group Preference in Network Security

随着电力通信网络及大数据应用技术的发展,用户对于网络安全的需求性也越来越高,因此网络中的用户喜好挖掘后的隐私保护也变得愈发重要。为了有针对性地实施终端用户隐私的保护,本文提出了一种基于用户偏好的主客观混合的体验质量(QoE)评估模型。本文以协同过滤方法为基础,根据用户使用记录实现基于用户群体的兴趣挖掘,从而满足单个用户数据的隐私保护并且降低模型判断复杂度,有效地保证了数据安全问题。实验结果表明,划分用户类型后的决策树可以获得更好的预测结果,保护了用户的隐私,从而保障了网络的安全。

With the development of power communication networks and big data application technologies, users are increasingly demanding network security. In the era of information, user interest mining and privacy protection become more important.In order to ensure the protection of privacy for end users, this manuscript proposes a subject-objective mixed quality of experience(Qo E) evaluation model based on service content and user interest. This manuscript incorporates the user group preference based on the collaborative filtering method, and realizes interest mining based on the user group according to the their using record, which satisfies the privacy protection of individual user data and reduces the complexity of model judgment. Based on the behavior pattern of user group, the extended program list is obtained to avoid single user data training, which effectively ensures the data security problem. Experimental results show that the decision tree after dividing the user type can obtain better prediction results and protect the privacy of users, thus ensure the security of the network.