An Improved Secure High-Order-Lanczos Based Orthogonal Tensor SVD for Outsourced Cyber-Physical-Social Big Data Reduction

Jun Feng, Laurence T. Yang, Guahui Dai, Jinjun Chen, Zheng Yan

Research output: Contribution to journalArticleScientificpeer-review

Abstract

Cyber-physical-social big data concern heterogeneous, multiaspect, large-volume data generated in cyber-physical-social systems (CPSS). Orthogonal tensor SVD has emerged as a powerful tool to reduce cyber-physical-social big data. In this work, we propose an improved secure high-order-Lanczos based orthogonal tensor SVD for cyber-physical-social big data reduction in clouds. Specifically, to take advantage of the parallel processing capability of cloud computing, the improved secure high-order Lanczos algorithm is derived by restructuring the original high-order Lanczos algorithm such that only one synchronization point per iteration is required. To protect data privacy, the improved secure high-order-Lanczos based orthogonal tensor SVD employs homomorphic encryption integrated with batching technique, and garbled circuits, and makes all computations of the orthogonal tensor SVD algorithm in clouds come true. This is, to our best knowledge, the first work to efficiently tackle big data reduction in clouds in a privacy-preserving manner. Finally, we prove that our improved approach is secure under the semi-honest model. And we evaluate the proposed improved secure orthogonal tensor SVD on real datasets. The results show that our proposed improved secure approach is efficient and scalable for cyber-physical-social big data reduction.
Original languageEnglish
JournalIEEE Transactions on Big Data
Early online date2018
DOIs
Publication statusE-pub ahead of print - 2018
MoE publication typeA1 Journal article-refereed

Keywords

  • Tensile stress
  • Big Data
  • Cloud computing
  • Data privacy
  • Synchronization
  • Encryption
  • Cyber-physical-social systems
  • privacy-preserving
  • encrypted data processing
  • tensor
  • high-order Lanczos
  • big data
  • cloud computing

Fingerprint

Dive into the research topics of 'An Improved Secure High-Order-Lanczos Based Orthogonal Tensor SVD for Outsourced Cyber-Physical-Social Big Data Reduction'. Together they form a unique fingerprint.

Cite this