The clustering co-efficient of my real network is zero. As a matter of fact, all the nodes of the network have a clustering coefficient of zero.

I am trying to compare my network to an ensemble of random networks I created using the Erdos-Renyi algorithm with the same number of nodes and edges as my real network. It says online that the clustering coefficient of a real network is much higher than a random network.

In my case, since the clustering coefficient of my network is zero, the situation is the opposite. What can I say about this phenomenon in my report. Any advice is very much appreciated.

  • $\begingroup$ Integer division error? 1/2=0 $\endgroup$ – Anony-Mousse Jun 3 '17 at 21:12
  • $\begingroup$ @Anony-Mousse I don't think so. The network has a total of 361 nodes and 695 edges. Cytoscape gives the value as zero and I think its highly unlikely that the software makes that kind of an error $\endgroup$ – The Last Word Jun 3 '17 at 21:15
  • $\begingroup$ @Anony-Mousse I could send you the adjacency matrix file of my network $\endgroup$ – The Last Word Jun 3 '17 at 21:17
  • $\begingroup$ Then you probably have a multimodal graph, if you don't get any triangles. Either way, a comparison with ER graphs then is misleading. $\endgroup$ – Anony-Mousse Jun 3 '17 at 21:18
  • 1
    $\begingroup$ Well, you need to first understand your network. If it is bimodal, you need to understand this. Then you can try to generate a random bimodal network for comparison. Obviously, the CC is useless on bimodal networks unless you project it onto its modes. $\endgroup$ – Anony-Mousse Jun 4 '17 at 7:09

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.