Quoting this paper - SmartWalk (https://dl.acm.org/doi/pdf/10.1145/2976749.2978319):
For graph privacy, strong link privacy relies on deep perturbation to the original graph, indicating a large random walk length. However, as the fixed random walk length increases, the perturbed graph gradually approaches to a random graph, incurring a significant loss of utility.
They propose a machine-learning based approach to determining the appropriate random walk length as a trade-off between utility and security/privacy. However, is there (at all) an anonymous or privacy-preserving method of conducting a random walk itself?
