Comments on: Review of “The Link Prediction Problem for Social Networks” http://kallander.net/will/blog/2008/05/22/review-of-the-link-prediction-problem-for-social-networks/ Random rumblings, rants, and reflections Tue, 06 Jan 2009 21:29:58 +0000 http://wordpress.org/?v=2.7 hourly 1 By: wkallander http://kallander.net/will/blog/2008/05/22/review-of-the-link-prediction-problem-for-social-networks/comment-page-1/#comment-89 wkallander Tue, 10 Jun 2008 09:16:44 +0000 http://kallander.net/will/blog/?p=27#comment-89 I just discovered this paper: Hasan, M., Chaoji, V., Salem, S., and Zaki, M. J., (2006). "Link Prediction using Supervised Learning," Workshop on Link Analysis, Counter-terrorism and Security (with SIAM Data Mining Conference), Bethesda, MD, April 2006. ...which discusses an attempt to use attribute data as I suggested in my review (they beat me to it!). It does occur to me however, that doing so sacrifices generality. Part of the beauty of Liben-Nowell and Kleinberg's approach is that it can be applied to any social network, without having to do feature selection. I just discovered this paper:
Hasan, M., Chaoji, V., Salem, S., and Zaki, M. J., (2006). “Link Prediction using Supervised Learning,” Workshop on Link Analysis, Counter-terrorism and Security (with SIAM Data Mining Conference), Bethesda, MD, April 2006.

…which discusses an attempt to use attribute data as I suggested in my review (they beat me to it!). It does occur to me however, that doing so sacrifices generality. Part of the beauty of Liben-Nowell and Kleinberg’s approach is that it can be applied to any social network, without having to do feature selection.

]]>