电子杂志虚拟阅读行为网络的建模与分析  

Modeling and Analysis of a Virtual Reading Behaviors Relationship Network Induced by a E-magazine

作　　者：刘杰[1,2] 赵志冲[1] 李先锋[1] 

机构地区：[1]武汉科技学院非线性科学研究中心,武汉430073 [2]武汉科技学院理学院,武汉430073

出　　处：《武汉理工大学学报》2010年第8期147-151,共5页Journal of Wuhan University of Technology

基　　金：湖北省统计局研究项目(HB092-25;CI:ZhaoZ-C/LiX-F)

摘　　要：对《科学网电子杂志》的博文栏目作者群因阅读行为构成的虚拟网络开展了相关研究。基于给定度序列随机化网络生成算法,建立了一个博文作者群因阅读行为而产生的无向、赋权的、虚拟社会网络数学模型。该模型能够比较真实地反映这一博文作者人群因杂志遴选博文共同发表而产生的复杂阅读行为;提出的"虚拟联系"构建方式融合了部分后验"调查统计规律",较已有相关研究中依照物理联系数据建模的方式更合理。通过细致的数值模拟、分析与比较发现:该类度"负相关"虚拟网络具有典型的小世界和无标度网络特征。提出的作者群"虚拟网络"的构建方式有助于分析此类高端虚拟舆论社区中观点的形成、传播,也使得识别和发现社区中意见领袖及成员的角色等更容易。This paper selects a virtual reading relationship network induced by the e-magazine＇s Blog Column of ScienceNet as a research object.Based on a newly proposed random network with given degree sequence generation algorithm,a mathematical network model is established and analyzed.This model can reflect the mechanism of such a virtual network and can be used to explain some complex behaviors of those Blogers＇ selective reading modes.This model seems more reasonable than some existing related studies methods by using physical contact data to model a relationship network,since the constructing way of virtual contacts integrated some statistics information.Numerical simulations further showed that,these negative degree correlation virtual networks are typical complex networks with both small-world and scale-free network characteristics.Methods proposed here can be used to analysis the information communication,the formation of public opinion,etc,and it also makes it easier to identify opinion leader and differ member role from each other in the virtual community on the web.

关 键 词：科学网电子杂志 虚拟网络 小世界网络 无标度网络 社区结构 

分 类 号：N949[自然科学总论—系统科学]