Skip to main content
. 2020 Nov 16;10:19848. doi: 10.1038/s41598-020-76860-2

Table 4.

The prediction accuracies of our proposed method as well as of the state-of-the-art methods we bench-marked it against, measured by AUC.

graphic file with name 41598_2020_76860_Tab4_HTML.jpg

Each experiment was executed 100 times with independent random network division of the training set and the probe set and the average value of all the 100 runs are reported.