School of Data and Computer Science, Sun Yat-sen University, Guangzhou 510006, China;
School of Information Science and Technology, Southwest Jiaotong University, Chengdu 610031, China.
Proc Natl Acad Sci U S A. 2018 Jul 17;115(29):7468-7472. doi: 10.1073/pnas.1710547115. Epub 2018 Jul 3.
Measuring and optimizing the influence of nodes in big-data online social networks are important for many practical applications, such as the viral marketing and the adoption of new products. As the viral spreading on a social network is a global process, it is commonly believed that measuring the influence of nodes inevitably requires the knowledge of the entire network. Using percolation theory, we show that the spreading process displays a nucleation behavior: Once a piece of information spreads from the seeds to more than a small characteristic number of nodes, it reaches a point of no return and will quickly reach the percolation cluster, regardless of the entire network structure; otherwise the spreading will be contained locally. Thus, we find that, without the knowledge of the entire network, any node's global influence can be accurately measured using this characteristic number, which is independent of the network size. This motivates an efficient algorithm with constant time complexity on the long-standing problem of best seed spreaders selection, with performance remarkably close to the true optimum.
测量和优化大数据在线社交网络中的节点的影响对于许多实际应用非常重要,例如病毒营销和新产品的采用。由于社交网络上的病毒传播是一个全局过程,因此人们普遍认为,测量节点的影响必然需要整个网络的知识。我们利用渗流理论表明,传播过程表现出成核行为:一旦一条信息从种子传播到超过一个小的特征数量的节点,它就会到达一个不可逆转的点,并且无论整个网络结构如何,它都会迅速到达渗流簇;否则,传播将在本地被包含。因此,我们发现,无需整个网络的知识,就可以使用这个特征数量准确测量任何节点的全局影响,而与网络大小无关。这激发了一种在长期存在的最佳种子传播者选择问题上具有恒定时间复杂度的高效算法,其性能与真实最优值非常接近。