Fingerprinting Big Data: The Case of KNN Graph Construction

Abstract : We propose fingerprinting, a new technique that consists in constructing compact, fast-to-compute and privacy-preserving binary representations of datasets. We illustrate the effectiveness of our approach on the emblematic big data problem of K-Nearest-Neighbor (KNN) graph construction and show that fingerprinting can drastically accelerate a large range of existing KNN algorithms, while efficiently obfuscating the original data, with little to no overhead. Our extensive evaluation of the resulting approach (dubbed GoldFinger) on several realistic datasets shows that our approach delivers speedups of up to 78.9% compared to the use of raw data while only incurring a negligible to moderate loss in terms of KNN quality. To convey the practical value of such a scheme, we apply it to item recommendation, and show that the loss in recommendation quality is negligible.
Complete list of metadatas

https://hal.inria.fr/hal-01904341
Contributor : Olivier Ruas <>
Submitted on : Wednesday, October 24, 2018 - 6:16:22 PM
Last modification on : Monday, December 3, 2018 - 10:20:04 PM
Long-term archiving on : Friday, January 25, 2019 - 3:41:46 PM

File

RR-9218.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01904341, version 1

Citation

Rachid Guerraoui, Anne-Marie Kermarrec, Olivier Ruas, François Taïani. Fingerprinting Big Data: The Case of KNN Graph Construction. [Research Report] RR-9218, INRIA Rennes - Bretagne Atlantique; INRIA - IRISA - PANAMA; Université de Rennes 1; EPFL; Mediego. 2018, pp.1-30. ⟨hal-01904341⟩

Share

Metrics

Record views

1023

Files downloads

156