Proving Ownership over Categorical Data

  • Authors:
  • Radu Sion

  • Affiliations:
  • -

  • Venue:
  • ICDE '04 Proceedings of the 20th International Conference on Data Engineering
  • Year:
  • 2004

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper introduces a novel method of rightsprotection for categorical data through watermarking.We discover new watermark embedding channelsfor relational data with categorical types. Wedesign novel watermark encoding algorithms andanalyze important theoretical bounds including markvulnerability. While fully preserving data qualityrequirements, our solution survives important attacks,such as subset selection and random alterations. Markdetection is fully "blind" in that it doesn't require theoriginal data, an important characteristic especiallyin the case of massive data. We propose variousimprovements and alternative encoding methods. Weperform validation experiments by watermarking theoutsourced Wal-Mart sales data available at ourinstitute. We prove (experimentally and by analysis)our solution to be extremely resilient to both alterationand data loss attacks, for example tolerating up to 80%data loss with a watermark alteration of only 25%.