Package jadex.commons.collection

Class BloomFilter

java.lang.Object

jadex.commons.collection.BloomFilter

public class BloomFilter
extends java.lang.Object

A bloom filter is a probabilistic data structure for checking if a value is contained in a set. It has a false positive rate, i.e. it can tell that a value is contained in the filter although it is not. This implementation does not support remove() because clearing a bit that is used by more than one element might lead to false negatives. (If remove is required a counting filter needs to be used).

Field Summary

Fields

Modifier and Type	Field	Description
protected java.util.BitSet	bs	The bit set.
protected int	k	The number of hash (functions).
protected int	m	The number of bits in the bit set
protected int	n	The number of expected entries.

Constructor Summary

Constructors

Constructor	Description
BloomFilter()	Create a new BloomFilter with default p = 0.01 and n = 1000.
BloomFilter(double p, int n)	Create a new BloomFilter.
BloomFilter(int m, int k, int n)	Create a new BloomFilter.

Method Summary

Modifier and Type	Method	Description
boolean	add(byte[] value)	Add a value to the filter.
void	clear()	Clear the filter content.
static int	computeAcceptableN(long k, long m)	Compute acceptable amount of elements with given number of hashes and size of filter in bits.
static int	computeOptimalK(long n, long m)	Compute optimal number of hash functions using expected number of elements and size of filter in bits.
static int	computeOptimalM(long n, double p)	Compute optimal size of bloom filter in bits by expected number of elements and acceptable false positive rate.
static double	computeP(long k, long m, double insertedElements)	Compute best-case (uniform hash function) false positive probability.
static int[]	hashK(byte[] value, int m, int k)	Compute k hashes based on two hashes using (hash1+i*hash2)%m;
static void	main(java.lang.String[] args)	Main for testing.
boolean	mightContain(byte[] value)	Test if a value is contained in the filter.
static int	murmur3(byte[] data, int seed)	Compute the murmur hash for a byte array.
java.lang.String	toString()	Get the string representation.

Methods inherited from class java.lang.Object

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

Field Detail

bs

protected java.util.BitSet bs

The bit set.

k

protected int k

The number of hash (functions).

m

protected int m

The number of bits in the bit set

n

protected int n

The number of expected entries.

Constructor Detail

BloomFilter

public BloomFilter()

Create a new BloomFilter with default p = 0.01 and n = 1000.

BloomFilter

public BloomFilter(double p,
                   int n)

Create a new BloomFilter.

Parameters:

p - The acceptable false positive rate.

n - The expected number of entries.

BloomFilter

public BloomFilter(int m,
                   int k,
                   int n)

Create a new BloomFilter.

Parameters:

m - The number of bits in the filter.

k - The number of hashes used.

n - The expected number of values in the filter.

Method Detail

add

public boolean add(byte[] value)

Add a value to the filter.

Parameters:

value - The value.

clear

public void clear()

Clear the filter content.

mightContain

public boolean mightContain(byte[] value)

Test if a value is contained in the filter.

Parameters:

value - The value.

Returns:

True, if value might be contained.

computeOptimalM

public static int computeOptimalM(long n,
                                  double p)

Compute optimal size of bloom filter in bits by expected number of elements and acceptable false positive rate.

Parameters:

n - Expected number of elements.

p - Acceptable false positive rate.

Returns:

The optimal size of the bloom filter in bits.

computeOptimalK

public static int computeOptimalK(long n,
                                  long m)

Compute optimal number of hash functions using expected number of elements and size of filter in bits.

Parameters:

n - Expected number of elements inserted in the bloom filter

m - The size of the bloom filter in bits.

Returns:

The optimal number of hash functions.

computeAcceptableN

public static int computeAcceptableN(long k,
                                     long m)

Compute acceptable amount of elements with given number of hashes and size of filter in bits.

Parameters:

k - Number of hashes.

m - The size of the bloom filter in bits.

Returns:

Acceptable number elements that can be inserted.

computeP

public static double computeP(long k,
                              long m,
                              double insertedElements)

Compute best-case (uniform hash function) false positive probability.

Parameters:

k - Number of hashes.

m - The size of the bloom filter in bits.

n - Number of elements inserted in the filter.

Returns:

The expected false positive probability.

hashK

public static int[] hashK(byte[] value,
                          int m,
                          int k)

Compute k hashes based on two hashes using (hash1+i*hash2)%m;

Parameters:

value - The byte array.

m - The maximum number of values.

k - The number of hash values to compute.

Returns:

K hashes.

murmur3

public static int murmur3(byte[] data,
                          int seed)

Compute the murmur hash for a byte array.

toString

public java.lang.String toString()

Get the string representation.

Overrides:

toString in class java.lang.Object

main

public static void main(java.lang.String[] args)

Main for testing.