Currently, there are some defects in data aggregation encryption methods, such as low accuracy and efficiency. In this regard, this paper proposes an aggregation encryption method of social network privacy data based on matrix decomposition. Shamir threshold segmentation scheme was used to improve the security of matrix decomposition results, and then re-encrypted network user attributes. For codomain and non-codomain users, private file and key generation technology were used to improve the privacy of users in common domain and non-common domain. Considering the results of matrix decomposition, privacy security and efficient user matching in social networks were implemented. The node information was embedded in the process of data encryption and signature, and the interference factors were classified by cluster as a unit, which can resist the internal attack and improve the operation efficiency. The experimental results show that this method has excellent encryption effect and high privacy data protection processing efficiency.

当前，数据聚合加密方法存在一些缺陷，例如准确性和效率低下。为此，本文提出了一种基于矩阵分解的社交网络隐私数据聚合加密方法。Shamir阈值分割方案用于提高矩阵分解结果的安全性，然后重新加密网络用户属性。对于共域和非共域用户，私有文件和密钥生成技术用于提高公共域和非公共域中用户的隐私。考虑到矩阵分解的结果，实现了社交网络中的隐私安全和有效的用户匹配。将节点信息嵌入到数据加密和签名过程中，并将干扰因素按群集分类为一个单元，可以抵御内部攻击，提高运作效率。实验结果表明，该方法具有优良的加密效果和较高的隐私数据保护处理效率。