Table 4 AUC for all the algorithms applied to real networks Hamsterster and Maayan-Vidal and their rewired versions
From: Influence of clustering coefficient on network embedding in link prediction
| Â | Â | Algorithms | |||
|---|---|---|---|---|---|
Datasets | Common Neighbours | Matrix Factorisation | Laplacian Eigenmaps | node2vec | |
Hamsterster | Original | \(0.749 \pm 0.003\) | \(\bf {0.807 \pm 0.005}\) | \(0.738 \pm 0.035\) | \(0.780 \pm 0.005\) |
\({{C_L = 0.20}}\) | \(0.788 \pm 0.004\) | \(\bf {0.806 \pm 0.004}\) | \(0.747 \pm 0.044\) | \(0.786 \pm 0.005\) | |
\({{C_L = 0.30}}\) | \(\bf {0.863 \pm 0.005}\) | \(0.800 \pm 0.006\) | \(0.777 \pm 0.040\) | \(0.823 \pm 0.007\) | |
Maayan- Vidal | Original | \(0.590 \pm 0.003\) | \(\bf {0.746 \pm 0.008}\) | \(0.624 \pm 0.016\) | \(0.607 \pm 0.008\) |
\({{C_L = 0.10}}\) | \(0.643 \pm 0.005\) | \(\bf {0.742 \pm 0.007}\) | \(0.636 \pm 0.016\) | \(0.617 \pm 0.008\) | |
\({{C_L = 0.20}}\) | \(\bf {0.744 \pm 0.006}\) | \(0.721 \pm 0.008\) | \(0.695 \pm 0.023\) | \(0.681 \pm 0.009\) | |