cqvip:伴随着网络技术的发展,各类社交网络所包含的信息也在不断地增大。在数据信息增加的同时也意味着隐私信息泄露的可能性增大。因此在上传和提取用户信息的时候应该考虑到敏感信息的保护,在k-匿名算法的基础上衍生的(a, k)-匿名算法是经典的隐私保护模型,但是随着社交网络的复杂性不断增加,传统的(a, k)-匿名算法不足以满足社交网络中信息隐匿的要求。针对在社交网络中,节点的结构信息和非隐私属性信息等也可能会受到攻击,本文提出一种基于节点分割的(a, k)-匿名算法。该算法对社交网络中带有隐私属性值的节点进行分割,使得节点特征被分割到两个节点里,降低了节点被攻击识别的可能性。实验结果表明,该算法可以有效防御部分攻击造成的隐私属性泄露,同时保证数据保持一定的可用性。With the development of network technology, the information contained in various social net-works is constantly increasing. But the increase in data information also means that the possibility of leakage of private information increases. Therefore, the protection of sensitive information should be considered when uploading and extracting user information. The (a, k)-anonymous al-gorithm derived from the k-anonymity algorithm is a classic privacy protection model, but with the complexity of social networks increasingly, the traditional (a, k)-anonymity algorithm is insuf-ficient to meet the requirements of information hiding in social networks. In social networks, structural information and non-privacy attribute information of nodes may also be attacked, in-creasing the risk of privacy attribute disclosure. A privacy attribute (a, k)-anonymous algorithm based on node segmentation is proposed in this paper. In this algorithm, the nodes with privacy attribute value in the social network are segmented, so that the features of the nodes are divided into two nodes, and the possibility of the nodes being attacked is reduced. Experimental results demonstrate that this algorithm can protect the privacy data from partial attacks and ensure the availability of data.
展开▼
