1. Convert a graph in webgraph format (dblp-2011.graph-txt) to a AD (Apostolico and Drovandi compression) format dblp-2011.net. 
   Both files are in text format

awk -F " " '{if(NR==1){printf("%d\n",$1);}else{printf("%d %s\n", NR-2, $0);}}' dblp-2011.graph-txt > dblp-2011.net

2. Execute wgraphToTxt to convert to the txt format I use
./wgraphToTxt dblp-2011.net > dblp-2011.txt

3. Execute txtTOBin to convert txt file to bin format. The bin file will be stored in the same directory of dblp-2011.txt with a name
dblp-2011.txt.bin. The first parameter is the number of nodes and the second the number of edges. Note that this is directed graph.

./txtTOBIN dblp-2011.txt 986324 6707236

4. Execute vnmextract. First create the output directory outdir
   
./vnmextract dblp-2011.txt.bin 1 1 100 500,100 outdir/dblp-2011 2   

This will create files with name dblp-2011-biclique-it-number.txt in the output directory outdir. Each output file has a line for each
biblique, where there are two sets of nodes separated by a caracter "-".  The set on the left side is the source set and the right set is the target
of each biclique. In addition, the output will write the graphs without bicliques in dblp-2011-it-p and dblp-2011-it-q, where p is the iteration where it stops extracting bicliques of size 500 and q the iteration where it stops extracting bicliques of size 100.
The size of a biclique is computed as number of nodes in source set multiplied by the number of nodes in target set.

With respecto to the parematers of vnmextract:

Do not change the 1 1 after the bin file. The next 100 indicates that the extraction will iterate in a biclique size until to the number of bicliques
in each biclique size is over 100. Be aware that if you specify the number of bibliques to be small it can take some time to extract the bicliques.
The 500,100 specifies the sizes of the bicliques you can to capture starting at 500. The last 2 indicates that it will use 2 hashes for minhashing.