Table 1 Comparison of prediction accuracy under the metric of Pearson correlation coefficient for the top-L ranked links.

Network\Index	CN	WCN	rWCN	AA	WAA	rWAA	RA	WRA	rWRA
Celegans	0.89	0.887	0.887	0.875	0.872	0.868	0.866	0.863	0.847
Celegans	0.964	0.965	0.964	0.965	0.965	0.964	0.965	0.966	0.966
Everglades	0.161	0.459	0.637	0.208	0.501	0.645	0.284	0.449	0.585
Everglades	0.858	0.793	0.81	0.85	0.87	0.815	0.84	0.882	0.812
USAir1	−0.0059	0.668	0.403	−0.0711	0.346	0.375	−0.0868	0.332	0.308
USAir1	0.531	0.636	0.557	0.481	0.592	0.54	0.382	0.507	0.556
USAir2	0.944	0.942	0.939	0.926	0.924	0.92	0.774	0.768	0.756
USAir2	0.995	0.995	0.995	0.995	0.995	0.995	0.994	0.994	0.994
Advogato	0.861	0.867	0.87	0.837	0.835	0.84	0.789	0.79	0.806
Advogato	0.963	0.963	0.963	0.962	0.961	0.962	0.956	0.956	0.957
Geom	0.808	0.863	0.876	0.816	0.853	0.878	0.855	0.853	0.871
Geom	0.958	0.961	0.964	0.942	0.948	0.957	0.919	0.919	0.941

In each network, the first row is the results achieved by the linear-correlation method, while the second row shows the accuracy of our method. Each accuracy value is an average over 100 independent random divisions of the links into a training set and a validation set.

Quick links

Search