Merge branch 'master' of https://github.com/tdebatty/java-LSH

Conflicts:
	src/main/java/info/debatty/java/lsh/LSHMinHash.java

Author: tdebatty

README.md

@@ -7,14 +7,16 @@ Locality Sensitive Hashing (LSH) is a family of hashing methods that tent to pro
 
 LSH functions have two main use cases:
 * Compute the signature of large input vectors. These signatures can be used to quickly estimate the similarity between vectors.
-* With a given number of buekcts, bin similar vectors together.
+* With a given number of buckets, bin similar vectors together.
 
 This library implements Locality Sensitive Hashing (LSH), as described in Leskovec, Rajaraman & Ullman (2014), "Mining of Massive Datasets", Cambridge University Press.
 
 Are currently implemented:
 * MinHash algorithm for Jaccard index;
 * Super-Bit algorithm for cosine similarity.
 
+The coeficients of hashing functions are randomly choosen when the LSH object is instantiated. You can thus only compare signatures or bucket binning generated by the same LSH object. To reuse your LSH object between executions, you have to serialize it and save it to a file (see below the [example of LSH object serialization](https://github.com/tdebatty/java-LSH#serialization)).
+
 ##Download
 
 Using maven:

src/main/java/info/debatty/java/lsh/LSHMinHash.java

@@ -24,6 +24,8 @@
 
 package info.debatty.java.lsh;
 
+import java.util.Set;
+
 /**
  *
  * @author Thibault Debatty