Research Article
The Influence of Network Structural Preference on Link Prediction
| (1) | Input: adjacent matrix of the original network , parameter , parameter , and parameter | | (2) | fordo | | (3) | DELETE-EDGE () | | (4) | end for | | (5) | function DELETE-EDGE () | | (6) | ; | | (7) | Initialize the value matrix: ; | | (8) | fordo | | (9) | fordo | | (10) | Calculate the value of each element in using formula (1); | | (11) | end for | | (12) | end for | | (13) | fordo | | (14) | Randomly choose a position based on the value matrix ; | | (15) | | | (16) | end for | | (17) | ; | | (18) | end function | | (19) | Output: the adjacent matrix after the perturbation has been applied |
|
