A simple implementation of simhash algorithm by java.

1. compute the simhash of a string
   
2. compute the similarity between all the strings by building smart index, so we can deal with big data.
   

• run Main with inputfile and outputfile.
  

• The format of inputfile(see src/test_in): one doc eachline with the utf8 charset.
  

• The format of outputfile(see src/test_out):
  

• start //start flag
  

• first line // doc
  

• sencode lien // doc1\tdist where dist is the hamming distance between doc and doc1
  

• end //end flag
  

1. Build the project to a runnable jar.
   
2. Improve the performace under big data.
   

1. Before run Main.java, you should choose a better analyzer instead of BinaryWordSeg!