PPMCK: Privacy-preserving multi-party computing for K-means clustering

Yongkai Fan, Jianrong Bai, Xia Lei, Weiguo Lin, Qian Hu, Guodong Wu, Jiaming Guo, Gang Tan

Research output: Contribution to journalArticlepeer-review

Abstract

The powerful resource advantage of the cloud provides a suitable computing environment for data processing. By transferring local computing to the cloud, the efficiency of data processing can be improved. However, the open cloud environment has defects in data privacy-preserving. In order to strengthen the protection of data privacy and ensure the security of multi-party interaction, we propose a privacy-preserving multi-party computing scheme for K-means clustering (PPMCK). PPMCK can preserve data privacy in the cloud and in the local side for each party from multi-party computing. In addition, PPMCK uses homomorphic encryption to protect data privacy. To support the division operation and ciphertext value size comparison with which homomorphic encryption cannot handle, the corresponding measurements are adopted, which make homomorphic encryption work smoothly. The experimental results demonstrate that PPMCK is effective in both data processing and privacy-preserving.

Original languageEnglish (US)
Pages (from-to)54-63
Number of pages10
JournalJournal of Parallel and Distributed Computing
Volume154
DOIs
StatePublished - Aug 2021

All Science Journal Classification (ASJC) codes

  • Software
  • Theoretical Computer Science
  • Hardware and Architecture
  • Computer Networks and Communications
  • Artificial Intelligence

Fingerprint

Dive into the research topics of 'PPMCK: Privacy-preserving multi-party computing for K-means clustering'. Together they form a unique fingerprint.

Cite this