Patent · US Active

Generation method and device for generating anonymous dataset, and method and device for risk evaluation

US9129117B2 · kind B2 · utility

5Cited by
3References
26Claims
0Family size

Assignee

Inventors

Key dates

Filing dateDec 27, 2012
Grant dateSep 8, 2015
Priority date
Expiry dateJun 19, 2033

Classification

  • Technology area (CPC G)Physics
  • CPC primaryG06F2221/034
  • WIPO fieldComputer technology
  • WIPO sectorElectrical engineering

Abstract

An anonymous dataset generation method comprises following steps. A critical attribute set and a quasi-identifier (QID) set are acquired, and one of the critical attribute and the quasi-identifier is set as an anchor attribute. An attribute sequence and an equivalence table are generated according to the quasi-identifier set and the critical attribute set. A data cluster and a cluster table are generated according to the equivalence table. The content of the cluster table is generalized to generate and output an anonymous dataset corresponding to an original dataset. A risk evaluation method for an anonymous dataset calculates data weight to extract distinctive data and to attacking defects of the anonymous dataset according to the distinctive data, thereby enhancing a risk evaluation efficiency of the anonymous dataset.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.