Sensitivity of community detection

[apitts]

Is it possible to pass an argument (or otherwise alter the code) such that a user can control the sensitivity of community detection (i.e. more sensitive would result in a larger number of communities and less sensitive would result in a smaller number)?

I see in the original paper the following: "However, our algorithm provides a decomposition of the network into communities for different levels of organization..... This result suggests that the intermediate solutions found by our algorithm may also be meaningful and that the uncovered hierarchical structure may allow the end-user to zoom in the network and to observe its structure with the desired resolution."

I am not clear though how I can obtain that flexibility / sensitivity. Thanks.

[upphiminn]

From what I remember the authors are referring to the in-between communities found while running the algorithm loop, you can recover those with the partition_at_level() method inside the lib, but I forgot to expose it to the outside world when I return the public methods. Is this what you require? It will return the community partition at a certain iteration of the algorithm.

[apitts]

Yes, that is it. Thanks! It might take me some time but if I am able to expose it and get it working I'll make a pull request.

[upphiminn]

I will try to expose it some time soon and make the update. Will let you know.

[grunfe]

This quick hack works for me. Example:

result[0]: 31 communities
result[1]: 17 communities

Without the patch the output would only be result[1]

--- jLouvain-orig.js 2015-02-25 14:07:08.766808283 +0000
+++ jLouvain.js        2015-02-25 14:11:38.358808283 +0000
@@ -339,7 +339,11 @@
                var core = function(){
                        var status = {};
                        var dendogram = generate_dendogram(original_graph, partition_init);
-                       return partition_at_level(dendogram, dendogram.length - 1);
+                       var result=[];
+                       for(var i=0; i<dendogram.length; i++) {
+                               result.push(partition_at_level(dendogram, i));
+                       }
+                       return result;
                };

                core.nodes = function(nds){

[apitts]

👍