Hi Gábor, > * Gábor Csárdi <[email protected]> [2012-08-14 > 11:39:07 -0400]: > > the spinglass methos has a resolution parameter (called gamma), so > tuning that should give you different results. Try to increase it from > the default 1. Hopefully spinglass won't take too long to run on your > data.
increasing it (even to 100) only gives me 25 communities with a very uneven distribution of members: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 269 234 252 241 136 67 135 165 631 139 29 82 250 1 377 318 17 18 19 20 21 22 23 24 25 575 51 61 115 171 168 1762 518 208 (for a highly pruned graph with v=6955 e=214234) I actually have vertex weights in addition to edge weights; is there a community detection algorithm which would try to make total community weights approximately the same? -- Sam Steingold (http://sds.podval.org/) on Ubuntu 12.04 (precise) X 11.0.11103000 http://www.childpsy.net/ http://dhimmi.com http://ffii.org http://memri.org http://camera.org http://www.memritv.org http://pmw.org.il Do not tell me what to do and I will not tell you where to go. _______________________________________________ igraph-help mailing list [email protected] https://lists.nongnu.org/mailman/listinfo/igraph-help
