Skip to main content

Advertisement

Springer Nature is making Coronavirus research free. View research | View latest news | Sign up for updates

Fig. 1 | Applied Network Science

Fig. 1

From: Entropy-based approach to missing-links prediction

Fig. 1

Comparison of link-prediction algorithms when applied to the undirected version of the WTW (top), the DIN (middle) and e-MID (bottom). Left panel: evolution of precision; middle panel: evolution of accuracy; right panel: evolution of AUC. Our entropy-based approach to link-prediction performs better than other algorithms, across the vast majority of temporal snapshots. Notice that PA implements the recipe \(s_{ij}^{PA}=k_{i}\cdot k_{j}\). See also Fig. 4 for a visual inspection of the errors accompanying the estimations shown here

Back to article page