摘要
通过对实际数据的分析,获得了短信网络的一些特点,包括:度分布和边权分布符合低头和重尾的幂律分布、平均点强度和点度不具有幂函数律的关联性等等。在此基础上,将用户间短信交互次数视为短信网络的边权,提出了一种加权短信网络演化模型,其中新节点加入采用加权局部优先连接机制,边权更新基于节点间亲密度及近期联系频繁度。仿真结果表明,该生成模型较好地符合了实际短信网络的统计特性。
By analyzing some factual data from short message service database, more features of short message networks （SMNs） are captured, including degree distribution as well as weight distribution demonstrating the behavior of power-law with droop-head and heavy-tail, average vertex weight with vertex degree without power-law behavior etc. And then, a weighted evolutionary model （WEM） is proposed for SMNs by taking the short message interactive times between users as weights. The proposed WEM adopts a weighted local priority mechanism for node growing and a weight updating scheme which is based on the familiarity and recent contact frequency between users. Simulation results show that WEM fits real SMNs pretty well.
出处
《电子科技大学学报》
EI
CAS
CSCD
北大核心
2012年第5期649-657,共9页
Journal of University of Electronic Science and Technology of China
基金
国家自然科学基金(61101136)
中芬国际科技合作项目(2010DFB10570)
关键词
近期联系频繁度
短信网络
加权局部优先
边权更新
recent contact frequency
SMS network
weighted local priority
weight updating
