EDDifferentialPrivacy (ARX API Documentation)

java.lang.Object
- org.deidentifier.arx.criteria.PrivacyCriterion
- - org.deidentifier.arx.criteria.ImplicitPrivacyCriterion
  - - org.deidentifier.arx.criteria.EDDifferentialPrivacy

All Implemented Interfaces:

java.io.Serializable
```
public class EDDifferentialPrivacy
extends ImplicitPrivacyCriterion
```
(e,d)-Differential Privacy implemented with SafePub as proposed in: Bild R, Kuhn KA, Prasser F. SafePub: A Truthful Data Anonymization Algorithm With Strong Privacy Guarantees. Proceedings on Privacy Enhancing Technologies. 2018(1):67-87. SafePub, in turn, is a practical implementation of (k,b)-SDGS which was originally proposed in: Ninghui Li, Wahbeh H. Qardaji, Dong Su: On sampling, anonymization, and differential privacy or, k-anonymization meets differential privacy. Proceedings of ASIACCS 2012. pp. 32-33

See Also:

Serialized Form

Constructor Summary

Constructors
Constructor and Description
`EDDifferentialPrivacy(double epsilon, double delta)` Creates a new data-dependent instance
`EDDifferentialPrivacy(double epsilon, double delta, DataGeneralizationScheme generalization)` Creates a new instance which is data-independent iff generalization is not null
`EDDifferentialPrivacy(double epsilon, double delta, DataGeneralizationScheme generalization, boolean deterministic)` Creates a new instance which may be configured to produce deterministic output.

Method Summary

All Methods Instance Methods Concrete Methods
Modifier and Type	Method and Description
`EDDifferentialPrivacy`	`clone()` Clone
`double`	`getBeta()` Returns the beta parameter of (k,b)-SDGS
`DataSubset`	`getDataSubset()` If a privacy model uses a data subset, it must overwrite this method
`double`	`getDelta()` Returns the delta parameter of (e,d)-DP
`double`	`getEpsilon()` Returns the epsilon parameter of (e,d)-DP
`DataGeneralizationScheme`	`getGeneralizationScheme()` Returns the defined generalization scheme
`int`	`getK()` Returns the k parameter of (k,b)-SDGS
`int`	`getMinimalClassSize()` If a privacy model provides a prosecutor risk threshold, it should override this method to enable optimizations
`int`	`getRequirements()` Returns the criterion's requirements.
`void`	`initialize(org.deidentifier.arx.framework.data.DataManager manager, ARXConfiguration config)` Override this to initialize the criterion.
`boolean`	`isAnonymous(org.deidentifier.arx.framework.lattice.Transformation<?> node, org.deidentifier.arx.framework.check.groupify.HashGroupifyEntry entry)` Implement this, to enforce the criterion.
`boolean`	`isDataDependent()` Returns whether this instance is data-dependent
`boolean`	`isDeterministic()` Returns whether this instance is deterministic
`boolean`	`isHeuristicSearchSupported()` Returns whether the criterion supports heuristic search strategies.
`boolean`	`isHeuristicSearchWithTimeLimitSupported()` Returns whether the criterion supports heuristic search strategies with time limits.
`boolean`	`isLocalRecodingSupported()` Returns whether the criterion supports local recoding.
`boolean`	`isMinimalClassSizeAvailable()` If a privacy model provides a prosecutor risk threshold, it should override this method to enable optimizations
`boolean`	`isOptimalSearchSupported()` Returns whether the criterion supports optimal search strategies.
`boolean`	`isSubsetAvailable()` If a privacy model uses a data subset, it must overwrite this method
`ElementData`	`render()` Renders the privacy model
`java.lang.String`	`toString()` Returns a string representation.

Methods inherited from class org.deidentifier.arx.criteria.PrivacyCriterion
clone, getPopulationModel, getRiskThresholdJournalist, getRiskThresholdMarketer, getRiskThresholdProsecutor, getSubset, isMonotonicWithGeneralization, isMonotonicWithSuppression, isSampleBased

Methods inherited from class java.lang.Object
equals, getClass, hashCode, notify, notifyAll, wait, wait, wait

- Constructor Detail
  - EDDifferentialPrivacy
```
public EDDifferentialPrivacy(double epsilon,
                             double delta,
                             DataGeneralizationScheme generalization)
```
    Creates a new instance which is data-independent iff generalization is not null
    
    Parameters:
    
    epsilon -
    
    delta -
    
    generalization -
  - EDDifferentialPrivacy
```
public EDDifferentialPrivacy(double epsilon,
                             double delta)
```
    Creates a new data-dependent instance
    
    Parameters:
    
    epsilon -
    
    delta -
  - EDDifferentialPrivacy
```
public EDDifferentialPrivacy(double epsilon,
                             double delta,
                             DataGeneralizationScheme generalization,
                             boolean deterministic)
```
    Creates a new instance which may be configured to produce deterministic output. Note: *never* set deterministic to true in production. This parameterization is for testing purposes, only.
    
    Parameters:
    
    epsilon -
    
    delta -
    
    generalization -
    
    deterministic -
- Method Detail
  - clone
```
public EDDifferentialPrivacy clone()
```
    Description copied from class: PrivacyCriterion
    
    Clone
    
    Specified by:
    
    clone in class PrivacyCriterion
  - getBeta
```
public double getBeta()
```
    Returns the beta parameter of (k,b)-SDGS
    
    Returns:
  - getDataSubset
```
public DataSubset getDataSubset()
```
    Description copied from class: PrivacyCriterion
    
    If a privacy model uses a data subset, it must overwrite this method
    
    Overrides:
    
    getDataSubset in class PrivacyCriterion
    
    Returns:
  - getDelta
```
public double getDelta()
```
    Returns the delta parameter of (e,d)-DP
    
    Returns:
  - getEpsilon
```
public double getEpsilon()
```
    Returns the epsilon parameter of (e,d)-DP
    
    Returns:
  - getGeneralizationScheme
```
public DataGeneralizationScheme getGeneralizationScheme()
```
    Returns the defined generalization scheme
    
    Returns:
  - getK
```
public int getK()
```
    Returns the k parameter of (k,b)-SDGS
    
    Returns:
  - getMinimalClassSize
```
public int getMinimalClassSize()
```
    Description copied from class: PrivacyCriterion
    
    If a privacy model provides a prosecutor risk threshold, it should override this method to enable optimizations
    
    Overrides:
    
    getMinimalClassSize in class PrivacyCriterion
    
    Returns:
  - getRequirements
```
public int getRequirements()
```
    Description copied from class: PrivacyCriterion
    
    Returns the criterion's requirements.
    
    Specified by:
    
    getRequirements in class PrivacyCriterion
    
    Returns:
  - initialize
```
public void initialize(org.deidentifier.arx.framework.data.DataManager manager,
                       ARXConfiguration config)
```
    Description copied from class: PrivacyCriterion
    
    Override this to initialize the criterion.
    
    Overrides:
    
    initialize in class PrivacyCriterion
    
    config - TODO
  - isAnonymous
```
public boolean isAnonymous(org.deidentifier.arx.framework.lattice.Transformation<?> node,
                           org.deidentifier.arx.framework.check.groupify.HashGroupifyEntry entry)
```
    Description copied from class: PrivacyCriterion
    
    Implement this, to enforce the criterion.
    
    Specified by:
    
    isAnonymous in class PrivacyCriterion
    
    Parameters:
    
    node - TODO
    
    Returns:
  - isDataDependent
```
public boolean isDataDependent()
```
    Returns whether this instance is data-dependent
    
    Returns:
  - isDeterministic
```
public boolean isDeterministic()
```
    Returns whether this instance is deterministic
    
    Returns:
  - isHeuristicSearchSupported
```
public boolean isHeuristicSearchSupported()
```
    Description copied from class: PrivacyCriterion
    
    Returns whether the criterion supports heuristic search strategies.
    
    Overrides:
    
    isHeuristicSearchSupported in class PrivacyCriterion
    
    Returns:
  - isHeuristicSearchWithTimeLimitSupported
```
public boolean isHeuristicSearchWithTimeLimitSupported()
```
    Description copied from class: PrivacyCriterion
    
    Returns whether the criterion supports heuristic search strategies with time limits.
    
    Overrides:
    
    isHeuristicSearchWithTimeLimitSupported in class PrivacyCriterion
    
    Returns:
  - isOptimalSearchSupported
```
public boolean isOptimalSearchSupported()
```
    Description copied from class: PrivacyCriterion
    
    Returns whether the criterion supports optimal search strategies.
    
    Overrides:
    
    isOptimalSearchSupported in class PrivacyCriterion
    
    Returns:
  - isLocalRecodingSupported
```
public boolean isLocalRecodingSupported()
```
    Description copied from class: PrivacyCriterion
    
    Returns whether the criterion supports local recoding.
    
    Specified by:
    
    isLocalRecodingSupported in class PrivacyCriterion
    
    Returns:
  - isMinimalClassSizeAvailable
```
public boolean isMinimalClassSizeAvailable()
```
    Description copied from class: PrivacyCriterion
    
    If a privacy model provides a prosecutor risk threshold, it should override this method to enable optimizations
    
    Overrides:
    
    isMinimalClassSizeAvailable in class PrivacyCriterion
    
    Returns:
  - isSubsetAvailable
```
public boolean isSubsetAvailable()
```
    Description copied from class: PrivacyCriterion
    
    If a privacy model uses a data subset, it must overwrite this method
    
    Overrides:
    
    isSubsetAvailable in class PrivacyCriterion
    
    Returns:
  - render
```
public ElementData render()
```
    Description copied from class: PrivacyCriterion
    
    Renders the privacy model
    
    Specified by:
    
    render in class PrivacyCriterion
    
    Returns:
  - toString
```
public java.lang.String toString()
```
    Description copied from class: PrivacyCriterion
    
    Returns a string representation.
    
    Specified by:
    
    toString in class PrivacyCriterion

Class EDDifferentialPrivacy

Constructor Summary

Method Summary

Methods inherited from class org.deidentifier.arx.criteria.PrivacyCriterion

Methods inherited from class java.lang.Object

Constructor Detail

EDDifferentialPrivacy

EDDifferentialPrivacy

EDDifferentialPrivacy

Method Detail

clone

getBeta

getDataSubset

getDelta

getEpsilon

getGeneralizationScheme

getK

getMinimalClassSize

getRequirements

initialize

isAnonymous

isDataDependent

isDeterministic

isHeuristicSearchSupported

isHeuristicSearchWithTimeLimitSupported

isOptimalSearchSupported

isLocalRecodingSupported

isMinimalClassSizeAvailable

isSubsetAvailable

render

toString