Table 4 The computation of link prediction methods.

From: Predicting missing links in complex networks based on common neighbors and distance

CN

sij = |Γ(i) ∩ Γ(j)|

Sal (Salton)

Jaccard

Sen (Sϕrensen)

HPI (Hub Promoted)

HDI (Hub Depressed)

LHN (Leicht-Holme-Newman)

AA (Adamic-Adar)

RA (Resource Allocation)

  

PA (Preferential Attachment)

sij = ki × kj

LP (Local Path)

sLP = A2 + βA3

  1. ki is the degree of node i. PA method and LP method do not directly relate to common neighbors, but based on local information, where A is adjacency matrix and β = 0.01.
Back to article page