Package dk.alexandra.fresco.stat.anonymisation

Class NoisyHistogram

Object

NoisyHistogram

All Implemented Interfaces:

Computation<List<DRes<SInt>>,ProtocolBuilderNumeric>

public class NoisyHistogram
extends Object
implements Computation<List<DRes<SInt>>,ProtocolBuilderNumeric>

Compute a differentially private histogram for a dataset.

Constructor Summary

Constructors

Constructor	Description
NoisyHistogram(List<DRes<SInt>> buckets, List<DRes<SInt>> data, double epsilon)	Given a list of upper bounds for buckets and a list of samples, this computation computes a differentially private histogram for the given buckets.

Method Summary

Modifier and Type	Method	Description
DRes<List<DRes<SInt>>>	buildComputation(ProtocolBuilderNumeric builder)

Methods inherited from class Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Details

NoisyHistogram

public NoisyHistogram(List<DRes<SInt>> buckets, List<DRes<SInt>> data, double epsilon)

Given a list of upper bounds for buckets and a list of samples, this computation computes a differentially private histogram for the given buckets. The last bucket contains all samples larger than the last upper bound.

Parameters:

buckets - Soft upper bounds for buckets

data - List of samples

epsilon - The privacy parameter

Method Details

buildComputation

public DRes<List<DRes<SInt>>> buildComputation(ProtocolBuilderNumeric builder)

Specified by:

buildComputation in interface Computation<List<DRes<SInt>>,ProtocolBuilderNumeric>