Skip to main content
Fig. 5 | Applied Network Science

Fig. 5

From: SURREAL: Subgraph Robust Representation Learning

Fig. 5

Performance sensitivity analysis of SURREAL. (a) Expansion process for |NE(u)|≥|NR(u)|=400. By increasing the size of NE(u), the corresponding Micro-F1 score decreases (b) Refinement process for |NR(u)|≤|NE(u)|=1200. By increasing the size of NR(u), the corresponding Micro-F1 score increases (c) Vary \(log^{d}_{2}\) from 4–9. Micro-F1 score increases up to \(log^{d}_{2}=7\), and then it starts to decrease

Back to article page