Efficient Privacy-Preserving k-Nearest Neighbor Search

Authors:
Yinian Qi;Mikhail J. Atallah
Affiliations:
-;-
Venue:
ICDCS '08 Proceedings of the 2008 The 28th International Conference on Distributed Computing Systems
Year:
2008

Citing 0
Cited 8

Privacy-Preserving Distributed k-Nearest Neighbor Mining on Horizontally Partitioned Multi-Party Data

ADMA '09 Proceedings of the 5th International Conference on Advanced Data Mining and Applications
Improved Garbled Circuit Building Blocks and Applications to Auctions and Computing Minima

CANS '09 Proceedings of the 8th International Conference on Cryptology and Network Security
3PRS: a personalized popular program recommendation system for digital TV for P2P social networks

Multimedia Tools and Applications
CPRS: A cloud-based program recommendation system for digital TV platforms

Future Generation Computer Systems
CPRS: a cloud-based program recommendation system for digital TV platforms

GPC'10 Proceedings of the 5th international conference on Advances in Grid and Pervasive Computing
SciMATE: A Novel MapReduce-Like Framework for Multiple Scientific Data Formats

CCGRID '12 Proceedings of the 2012 12th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (ccgrid 2012)
Secure k-NN computation on encrypted cloud data without sharing key with query users

Proceedings of the 2013 international workshop on Security in cloud computing
Secure k-NN query on encrypted cloud database without key-sharing

International Journal of Electronic Security and Digital Forensics

Quantified Score

Hi-index	0.00

Visualization

Abstract

We give efficient protocols for secure and private k-nearest neighbor (k-NN) search, when the data is distributed between two parties who want to cooperatively compute the answers without revealing to each other their private data. Our protocol for the single-step k-NN search is provably secure and has linear computation and communication complexity. Previous work on this problem had a quadratic complexity, and also leaked information about the parties' inputs. We adapt our techniquesto also solve the general multi-step k-NN search, and describe a specific embodiment of it for the case of sequence data. The protocols and correctness proofs can be extended to suit other privacy-preserving data mining tasks, such as classification and outlier detection.