Table 1 Description of PPDM methods
From: A comprehensive review on privacy preserving data mining
PPDM methods | Description |
|---|---|
Data distribution | May contain vertically or horizontally partitioned data |
Data distortion | Contain perturbation, blocking, aggregation or merging, swapping and sampling |
Data mining algorithms | Encloses classification mining, association rule mining, clustering, and Bayesian networks etc |
Data or rules hidden | Denotes to hide main data or rules of innovative data |
K-anonymity | Achieve the anonymization |
L-diverse | Keeps the least group size of K, and maintains the diversity of the sensitive attributes |
Taxonomy Tree | Attributes the generalization to limit the information leakage |
Randomization | An un-sophisticated and valuable technique to hide the individual data in PPDM |
Privacy protection | Protects the privacy, it should adapt data carefully to attain optimum data utility |