jshun / ppopp20-ae Goto Github PK

Hi Julian,

I am trying to convert SNAP graph to Hygra format using the Hygra conversion scripts, but the generated Hygra file is different from the one downloaded directly from Amazon AWS.

The SNAP graph I am testing is "com-orkut", which I downloaded from SNAP and the file is called "com-orkut.ungraph.txt.gz".

I followed this command, "./communityToHyperAdj com-orkut.ungraph.txt unweight-com-orkut-hygra", to generate the unweight Hygra file as follows.

$head unweight-com-orkut-hygra
AdjacencyHypergraph
3072441
234370166
117185083
234370166
0
12
124
165
232

which is different from the unweight one downloaded from AWS, which looks as follows.

$head com-orkut-hygra
AdjacencyHypergraph
2322299
107080530
15301901
107080530
0
8
11
22
29

I also generated the weighted one using "./communityToHyperAdj -w com-orkut.ungraph.txt weight-com-orkut-hygra", which looks like this:

$head weight-com-orkut-hygra
WeightedAdjacencyHypergraph
3019699
175777625
58592541
175777625
0
58592543
58592549
58592606
58592626

It is different from the weighted one downloaded, which is shown below.

$head com-orkut-wgh-hygra
WeightedAdjacencyHypergraph
2322299
107080530
15301901
107080530
0
8
11
22
29

Could you please check and let me know if I made any mistakes in generation? Thanks.

Also, I am confused about the specification of the input SNAP graph. For example, the SNAP com-orkut graph, which looks like this.

$head com-orkut.ungraph.txt
# Undirected graph: ../../data/output/orkut.txt
# Orkut
# Nodes: 3072441 Edges: 117185083
# FromNodeId ToNodeId
1 2
1 3
1 4
1 5
1 6
1 7

My question is that, is this a clique expansion representation, meaning the first column is a vertex and the second column is also a vertex? Or is it a bipartite representation, meaning the first column is a vertex while the second column is a hyperedge node?

Thank you very much for your help.

Lenny