This is a course project for Graph Analytics, inspired and based on paper Estimating properties of social networks via randomwalk considering private nodes.
We first give an review and analysis of their proposing estimators, and propose an improved estimator for average clustering coefficient with less error following similar ideas. At last, we evaluate performance of these estimators on various datasets and analysis the results.