1: \begin{abstract}

2:     Graphlets are induced subgraph patterns and  have been frequently applied to

3:     characterize the local topology structures of graphs across various

4:     domains, e.g., online social networks (OSNs) and biological networks.

5:     Discovering and computing graphlet statistics are highly challenging.

6:     First, the massive size of real-world graphs makes the exact computation of

7:     graphlets extremely expensive.  Secondly, the graph topology may not be

8:     readily available so one has to resort to web crawling using the available

9:     application programming interfaces (APIs).

10:     In this work, we propose a general and novel framework to estimate graphlet

11:     statistics of ``{\em any size}''. Our framework is based on collecting

12:     samples through consecutive steps of random walks. We derive an analytical

13:     bound on the sample size (via the Chernoff-Hoeffding technique) to

14:     guarantee the convergence of our unbiased estimator. 

15:     To further improve the accuracy, we introduce two novel optimization techniques

16:     to reduce the lower bound on the sample size.  Experimental evaluations

17:     demonstrate that our methods outperform the state-of-the-art method up to

18:     an order of magnitude both in terms of accuracy and time cost.

19: \end{abstract}

20: