K-Neighborhood Shortest Path Privacy in the Cloud

Abstract

Preserving privacy on various forms of published data has been studied extensively in recent years. In particular, shortest distance computing in the cloud, while maintaining neighborhood privacy, attracts latest attention. To preserve fixed-pattern one-neighborhood privacy, current approach requires the calculation of all-pairs shortest paths in advance, which is time consuming for large graphs. In this work, we propose a new flexible k-neighborhood privacy-protected and efficient shortest distance computation scheme in the cloud. Combining k-skip shortest path sub-graphs, vertex hierarchy labeling and bottom-up partitioning, the proposed technique not only subsumes one-neighborhood privacy but also provides efficient partitioning and query processing. Numerical experiments demonstrating the characteristics of proposed approach are presented.