INPUT: |
the data table need to be published: PT; |
the anonymous parameter: ; |
Quasi-identifiers: ; |
OUTPUT: |
the data table GT satisfies -anonymity; |
the list of quasi-identifiers: LF; |
the list of records with similar quasi-identifiers: F; |
the list LO as the list of L; |
Repeat |
Calculate between and other records ; |
(3)Cluster the nearest records with ; |
(4)Remove these records form LF, and add these records to L; |
(5)Select the farthest records from to be the new core; |
(6)Add L to LO, then clear L. |
(7)untilāā |
(8)The remained records in LF are assigned to the nearest cluster in LO; |
(9)Unify each cluster in LO by Generalizing; |
(10)Create GT, QI from LO, and create SD from PT; |
(11)return GT |