info.debatty.java.lsh

Class LSH

java.lang.Object

info.debatty.java.lsh.LSH

All Implemented Interfaces:

Serializable

Direct Known Subclasses:

LSHMinHash, LSHSuperBit

public abstract class LSH
extends Object
implements Serializable

Implementation of Locality Sensitive Hashing (LSH) principle, as described in Leskovec, Rajaraman & Ullman (2014), "Mining of Massive Datasets", Cambridge University Press.

Author:

Thibault Debatty http://www.debatty.info

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
protected static long	LARGE_PRIME

Constructor Summary

Constructors

Constructor and Description
LSH() Instantiate an empty LSH instance (useful only for serialization).
LSH(int stages, int buckets) Instantiates a LSH instance with s stages (or bands) and b buckets (per stage), in a space with n dimensions.

Method Summary

Modifier and Type	Method and Description
int[]	hashSignature(boolean[] signature) Hash a signature.
int[]	hashSignature(int[] signature) Hash a signature.

Methods inherited from class java.lang.Object

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

Field Detail

LARGE_PRIME

protected static final long LARGE_PRIME

See Also:

Constant Field Values

Constructor Detail

LSH

public LSH(int stages,
           int buckets)

Instantiates a LSH instance with s stages (or bands) and b buckets (per stage), in a space with n dimensions.

Parameters:

stages - stages

buckets - buckets (per stage)

LSH

public LSH()

Instantiate an empty LSH instance (useful only for serialization).

Method Detail

hashSignature

public final int[] hashSignature(int[] signature)

Hash a signature. The signature is divided in s stages (or bands). Each stage is hashed to one of the b buckets.

Parameters:

signature -

Returns:

An vector of s integers (between 0 and b-1)

hashSignature

public final int[] hashSignature(boolean[] signature)

Hash a signature. The signature is divided in s stages (or bands). Each stage is hashed to one of the b buckets.

Parameters:

signature -

Returns:

An vector of s integers (between 0 and b-1)