基于社团并行发现的在线社交网络蠕虫抑制

Parallel Community Detection Based Worm Containment in Online Social Network

随着在线社交网络(Online Social Network,OSN)的快速发展,OSN蠕虫已经成为最具威胁的网络安全问题之一.为了防止OSN蠕虫的快速传播,文中提出了一种基于社团并行发现的OSN蠕虫抑制方法.首先将分布式图计算框架Pregel和基于标签传播的社团发现算法(Label Propagation Algorithm,LPA)相结合,提出了一种能够处理大规模OSN网络社团发现问题的并行LPA算法(Parallel LPA,PLPA).其次,文中在PLPA算法的基础上给出了3种社团关键节点的选取策略,并提出了相应的OSN蠕虫抑制方法.最后,通过在两组真实数据集上进行的社团并行发现及OSN蠕虫抑制仿真实验证明了文中方法的有效性.

With the rapid development of Online Social Networks （OSNs）, worms propagating in these networks have become one of the most threatening security problems. To contain these rapidly spreading worms, in this paper, we propose a defensive measure which is based on parallel community detection in OSNs. Specifically, according to the Pregel data-processing infrastructure, we implement a new parallel version of label propagation algorithm （PLPA） that is capable of quickly finding communities in OSNs owning millions of users. And then we give three definitions for the influential users to whom we will first distribute patches to contain the propagation of OSN worms. To evaluate the performance of our approaches we test them on our simulating framework with two large-scale OSN datasets and analyze the experimental results which can show the effectiveness of our approaches.