Table 2 Similarity measures for link prediction
From: Graph-based machine learning model for weight prediction in protein–protein networks
Similarity measure | Formula | Path length | Weighted? |
|---|---|---|---|
Common neighbors (CN) | \(CN(X,Y)=|N(X)\cap N(Y)|\) | 2 | No |
Jaccard coefficient (CJ) | \(CJ(X,Y)=\frac{CN(X,Y)}{TN(X,Y)}\) | 2 | No |
Adamic adar (AA) | \(AA(X,Y)=\sum _{u\in N(X)\cap N(Y)}\frac{1}{\log |N(u)|}\) | 2 | No |
Preferential attachment (PA) | \(PA(X,Y)=|N(X)|*|N(Y)|\) | 2 | No |
Resource allocation (RA) | \(RA(X,Y)=\sum _{u\in N(X)\cap N(Y)}\frac{1}{|N(u)|}\) | 2 | No |
Length-3 paths (L3) | \(P(X,Y)=\sum _{U,V}\frac{a_{XU}*a_{UV}*a_{VY}}{\sqrt{|N(U)||N(V)|}}\) | 3 | No |