Abstract:

The development of cloud computing has brought about rapid changes in the data application environment. The contradiction between the availability of data fusion and privacy protection is increasing. Traditional information security focuses on hiding attributes, while big data privacy protection pays more attention to access control and data using methods. As a semi-structured representation model, XML data is an ideal carrier for heterogeneous data exchange and unstructured data storage. This paper deeply studies several problems involved in user's individual access process for XML data privacy preserving, and proposes prior knowledge as dynamic context. Based on the innovative privacy bipartite graph and the XML document semantic encoding, we propose the XML data privacy preserving model DCPPM and the algorithm KCQ to resist reasoning attack. The scheme to preventive response of the real-time inference attack in the actual operation process is realized. Finally, the example verification and experimental data show that the proposed model can effectively protect the safe use of private data and sensitive data in real time, which has certain theoretical and practical significance.

Key words: dynamic context, reasoning attack, privacy preserving, privacy bipartite graph, LT-BT encoding, TS encoding