项目名称: 网络重要节点及链路挖掘方法研究
项目编号: No.11205042
项目类型: 青年科学基金项目
立项/批准年度: 2013
项目学科: 物理学II
项目作者: 吕琳媛
作者单位: 杭州师范大学
项目金额: 22万元
中文摘要: 复杂网络是交叉科学研究的重大热点,其中如何挖掘网络中的重要节点和链路受到越来越广泛的关注。这方面的研究不仅能够在微观层面上更加细致的剖析网络的结构特征,而且能够作为研究和分析网络动力学的辅助工具。同时,重要节点和链路的挖掘还能够产生直接的社会经济价值。本项目拟将统计物理的思路、理论和方法引入来解决节点和链路挖掘的问题,并利用相关结果分析网络动力学,实现从理论、算法、应用三个层面全方位推动相关研究。本项目将利用网络结构信息设计快速有效的重要节点/链路挖掘方法,针对不同的网络形式和排序目标讨论不同算法挖掘出的节点/链路对网络特定结构和功能的影响。特别关注多层网络上关键节点/链路的挖掘,区分链路的不同角色的方法,以及利用功能表征挖掘重要节点的反问题。本项目拟利用现有互联网和手机运营平台,设计和实现真实实验,获取无控制的真实行为数据,进行实验分析,检验和丰富理论研究的结果,并指导应用开发。
中文关键词: 复杂网络;节点中心性;排序算法;链路预测;网络演化
英文摘要: Studies on complex networks become one of the most exciting domains of interdisciplinary researches. Identifying influential nodes and important links of networks attracts much attention, since it has both theoretical and practical significance. Research on this area can not only help to reveal microscopic structures of nerworks, but also provide assistant tools on analyzing network dynamics. Meanwhile, it can find real applications with great social and economic values. In this project, we will apply perspectives, theories and methods in statistical physics to find out the important nodes and links on networks, and further analyze their effects on network dynamics. With different ranking purposes, the corresponding algorithms will be designed based on the network structure. Especially, we will focus on the problem on multi-level networks, distinguish the different roles of nodes and links, and discuss how to utilize the features of network functions to identify the important nodes and links. Finally, we will conduct real experiments on WWW and mobile networks to test the algorithms. By analyzing the real behavior data, we can further improve the algorithms and develop applications in real systems.
英文关键词: complex networks;node centrality;ranking method;link prediction;network evolution