Privacy-Preserving Top-k Nearest Keyword Search on Outsourced Graphs

摘要

With massive networks emerging with labels or textual contents on the nodes, keyword search on graphs has been used in a wide range of real-life applications in recent years. For achieving great cost savings, data owners are motivated to outsource their services on graphs to the cloud. However, directly outsourcing them may arise serious privacy concerns. In this paper, we study the problem of privacy-preserving top-k nearest keyword search on outsourced graphs. Only a few existing studies primarily focus on the privacy-preserving graph operations under encryption settings, which cannot be directly applied to solve the problem of privacy-preserving keyword search on graphs. To address this problem, we propose a new privacy-preserving scheme for top-k nearest keyword search on graphs, in which a two-level secure index is devised to facilitate privacy-preserving top-k nearest keyword search. To handle the keyword filtering in search processing, we also propose a trapdoor generation method based on privacy-preserving set operations. Leveraging the two-level secure index and trapdoors, we further present the privacy-preserving top-k nearest keyword search algorithm. Thorough analysis shows the validity and security of our scheme. Extensive experimental results on real datasets further demonstrate our scheme can achieve high efficiency.