Science Advances

Supplementary Materials

This PDF file includes:

  • Supplementary Materials and Methods
  • fig. S1. Belief propagation method used in our approach.
  • fig. S2. Topological effects on the accuracy of the model.
  • fig. S3. Dependence of the algorithm performance on perturbation size and fraction of observed nodes.
  • fig. S4. Dependence of the performance of our approach versus the fraction of perturbed nodes in the observation.
  • fig. S5. Expectation maximization-estimated model parameter η* versus perturbation parameter c.
  • fig. S6. Accuracy of a k-neighbor classifier at detecting perturbations.
  • fig. S7. Perturbations on synthetic networks.
  • References (4244)

Download PDF

Files in this Data Supplement: