基于中心性的突发事件信息传播超网络中重要节点的评判研究

VIP免费

3.0 赵德峰 2024-11-19 4 4 1.97MB 114 页 15积分

侵权投诉

摘要

近年来突发事件发生较多，随着信息技术的快速发展和社会系统的日益复

杂，与突发事件相关的信息传播也越来越快速、隐蔽，如果我们不能及时了解相

关舆情或谣言在网络中的传播规律，并对其进行有效的引导与管理，任其扩散爆

发，将可能造成灾难性的后果。从现有研究看，较多工作偏重于同属性节点网络

中的信息传播研究，然而，对突发事件信息传播系统而言，同质信息传播网络模

型具有很大的片面性和局限性。作为一个复杂的社会系统，突发事件信息传播系

统中涵括了多种不同类别的要素，在同质或异质要素之间还存在着复杂的相互关

联。如此复杂的结构形态和构成，很难用单一要素模式的网络模型进行描述，更

难以进行全面深入的分析与探究。另外，从突发事件信息传播管理的研究及应用

来看，也迫切需要建立较为真实的、具有代表性的、可较全面深入描述突发事件

信息传播系统的整体集成模型，而同质节点的突发事件信息传播网络无法满足这

种要求，因此，只有在异质节点的信息传播网络思想基础上进一步深入研究才有

可能。而全面、合理地评判出突发事件信息传播系统中的重要节点则在突发事件

信息传播的相关研究中具有十分重要的意义与价值。首先，评判的结果可以用来

干预传播，如加速有重要社会价值信息的传播或监督防控谣言的扩散；其次，对

重要节点的分析，有助于我们理解突发事件信息传播网络的演化以及突发事件信

息传播的模式与途径；最后，这些研究的结果可以考虑应用于社会推荐和广告投

放等领域。基于上述情况，本文将超网络和中心性的理论方法综合应用于突发事

件信息传播系统中重要节点的评判研究，以期能够更加的全面与合理。

本文的工作及创新主要体现在以下几个方面：

（1）将超网络的理论和方法应用于突发事件的信息传播研究，对突发事件信

息传播超网络进行建模与分析。在突发事件的信息传播系统中，以人、互联网信息

载体、手机这三种相对较为活跃和较具代表性的信息载体为例，分别针对这三种信

息载体建立三种类型的信息传播网络，并根据三者间的关联关系将它们集成为一个突

发事件信息传播系统的“超网络”。在此基础上，对超网络模型的子网差异性、超网

络模型的特性及应用进行分析与探讨，并通过例子建立超网络模型的邻接矩阵与网络

直观图。

（2）基于程度中心性的相关理论方法对突发事件信息传播超网络中的重要节

点进行评判研究。节点的关联程度主要是指在突发事件信息传播系统中节点与信息

传播渠道、信息传播路径的关联情况，一个节点关联的信息传播渠道、信息传播路

径越多，则其可以进行信息传播的能力也就越大。程度中心性方法是当前最常用的

节点关联程度评判方法之一，有着广泛的应用，但在超网络视域下，该方法却存在

一定的局限性，依据现有算法的不足，提出一种基于程度中心性的突发事件信息传

播超网络中重要节点的评判方法，以实现对节点关联程度的评判更加全面与合理。

（3）基于中间中心性的相关理论方法对突发事件信息传播超网络中的重要节

点进行评判研究。突发事件信息传播系统中的调控节点即系统中能够介入和控制其

他节点之间交往和信息传递的节点，此类节点的关联程度未必较高，信息传播能力

未必较强，但其对突发事件信息传播进程及效果的影响巨大，因而在突发事件的信

息管理中对该类节点做出全面、合理的评判非常必要，利于监测预警、引导管控等

工作的开展与优化。本文将超网络和中间中心性的理论方法应用于突发事件信息传

播系统中重要调控节点的评判研究，构建一个基于中间中心性的突发事件信息传播

超网络中重要节点的综合评判方法，以实现对节点调控能力的评判更加全面与合

理。

（4）基于接近中心性的相关理论方法对突发事件信息传播超网络中的重要节

点进行评判研究。网络中的某个节点若在信息传播的过程中较少地依赖于其他节点，

则此节点就具有较高的枢纽地位，一个非枢纽位置的节点必须通过其他节点才能传递

信息。因此，应该考虑某节点与其它节点的接近程度，一个节点越是与其它节点接

近，该节点就越不依赖于其它节点，而且一个节点越是与其它节点接近，该节点在传

递信息方面也更为快速和容易。接近中心性是一种对节点不受其他节点控制的能力、

快捷传递信息的能力和节点枢纽地位的测度。虽然接近中心性方法在计算节点枢纽地

位方面较为准确和兼顾整体，但其也存在着一些不足。依据现有算法的不足，提出一

种基于接近中心性的突发事件信息传播超网络中重要节点的综合评判方法，并通过

例子验证分析了新方法的可行性与有效性。

（5）基于特征向量中心性的相关理论方法对突发事件信息传播超网络中的重

要节点进行评判研究。特征向量中心性算法理论认为一个节点的重要程度与其相连

其他节点的重要程度息息相关，即对于一个节点来说，如果该节点与很多自身具有

较高重要程度的节点相连接的话，那么该节点就具有较高的重要程度。本文将超网

络和特征向量中心性的理论方法综合应用于突发事件信息传播系统中重要节点的评

判研究，提出一种基于特征向量中心性的突发事件信息传播超网络中节点信息传播

综合能力的评判方法，并通过例子验证分析了新方法的可行性与有效性。

关键词：突发事件信息传播超网络程度中心性中间中心性接

近中心性特征向量中心性

ABSTRACT

In recent years, the emergencies happen frequently. Along with the complication of

the social system and the rapid development of information technology, the

emergencies’ information dissemination is also more covert and fast. If we cannot find

out the dissemination of the relevant public opinion or rumours on the network

opportunely and carry on effective guidance and management, the information eruption

may cause disastrous consequences. Most of the current study mainly focuses on the

same attributes nodes’ information dissemination. However, for the emergencies’

information dissemination system, the homogeneous information dissemination network

models have a lot of one-sidedness and limitations. As a complex social system,

emergencies’ information dissemination system contains a variety of different elements,

and the complex interrelations among them. It is therefore difficult to use a network of

single element model to describe such a complex structure and morphology, and more

so to carry out a comprehensive and in-depth analysis and research about it. In addition,

to study and apply emergency information dissemination management, it is also urgent

to establish the overall integrated model. The model should be more real, representative

and can present a more comprehensive and in-depth description of the emergencies’

information dissemination system. However, the homogeneous nodes’ networks of the

emergencies’ information dissemination can’t meet this requirement. Therefore, only on

the basis of further research on the heterogeneous nodes’ information dissemination

network idea, the establishment of a suitable model is possible. The comprehensive and

rational determination of the important nodes in the emergencies’ information

dissemination system is of vital significance. In the first place, the results can be used to

interfere in the information dissemination, for example, to accelerate the dissemination

of information of important social value or to monitor or control rumor’s dissemination.

In the second place, the analysis of the important nodes contributes to our understanding

of the evolution of emergencies’ information dissemination network and the mode and

pathway of emergencies’ information dissemination. Finally, the results of the studies

may be applied to social recommendation and advertising. Based on these, this paper

applies the theory of supernetwork and centrality to the determination of the important

nodes in the emergencies’ information dissemination system in order to get the

comprehensive and rational results.

The innovation and achievements of this dissertation mainly reflects in the

following five aspects:

(1) This paper applies the theory of supernetwork to the emergencies’ information

dissemination research. It presents the modeling and analysis of the emergencies

information dissemination supernetwork. The mobile phone, person and Internet

information carrier are the most active and representative elements in the emergencies

information dissemination system. This paper, based on the theory of supernetwork,

presents a supernetwork model which comprises mobile phone network, the personnel

network and the Internet information carrier network, and the complex interrelations

among them. Based on these, this paper analyses the model’s sub-network differences,

the model’s characteristics and the model’s application. It builds the adjacency matrix

and the visual network graph of the supernetwork through an example.

(2) This paper, based on the theory of degree centrality, researches the

determination of the important nodes in the emergencies information dissemination

supernetwork. The correlation degree of a node is mainly refers to the number of

information dissemination channels that a node has in the emergencies information

dissemination system. If a node correlates many channels of information dissemination,

the information dissemination ability of this node is strong. Degree centrality algorithm is

popular in the determination of the node’s correlation degree. But this method has some

limitations from the perspective of supernetwork. Based on the lack of existing method,

this paper proposes a new algorithm based on degree centrality to determine the important

nodes from the perspective of supernetwork in order to get the comprehensive and

rational results.

(3) Based on the theory of betweenness centrality, this paper researches the

determination of the important nodes in the emergencies information dissemination

supernetwork. The control nodes can intervene and control the other nodes’

communication and information transfer in the emergencies’ information dissemination

system. The information dissemination ability of these nodes may not be strong. But they

can greatly control or influence emergencies’ information dissemination process and

information dissemination effect. Thus the comprehensive and rational determination of

these nodes is very necessary in the emergencies’ information dissemination. It can

facilitate the actualizing and optimization of the monitoring and early warning, guidance

and control. This paper, based on the theory of supernetwork and betweenness centrality,

mainly researches the determination of the important control nodes in the emergencies’

information dissemination. In order to get the comprehensive and rational results, this

paper proposes a new algorithm based on the betweenness centrality to determine the

important nodes in the emergencies information dissemination supernetwork.

(4) This paper, based on the theory of closeness centrality, researches the

determination of the important nodes in the emergencies information dissemination

supernetwork. A central node is one that is not dependent upon others as intermediaries of

messages. A non-central node is one that must relay messages through others. A node is close

to the other nodes, the node is not dependent on other nodes. A node is close to the other

nodes, the node can easily communicate with other members of the network. These central

nodes can avoid the control potential of others. These central nodes can easily communicate

with other members of the network. Closeness centrality is a measure of these central nodes.

Closeness centrality algorithm is popular in the determination of the central nodes. But this

method also has some limitations. Based on the lack of existing algorithm, this paper

proposes a new algorithm based on the closeness centrality to determine the important

nodes in the emergencies information dissemination supernetwork. Lastly, the new

algorithm is illustrated through an example.

(5) Based on the theory of eigenvector centrality, this paper researches the

determination of the important nodes in the emergencies information dissemination

supernetwork. In many circumstances the value of a node in a network is a function of the

values of the nodes to which it is connected. If a node connects many important nodes, the

node is important. This paper, based on the theory of supernetwork and eigenvector

centrality, mainly researches the determination of the important nodes in the emergencies

information dissemination system. This paper, proposes a new method based on

eigenvector centrality algorithm to determine the important nodes from the perspective of

supernetwork. Finally, the new method is illustrated through an example.

Key Words: emergency, information dissemination, supernetwork,

degree centrality, betweenness centrality, closeness centrality,

eigenvector centrality

中文摘要

ABSTRACT

第一章绪论.................................................................................................................. 1

§ 1.1 课题的来源及意义 ............................................................................................ 1

§ 1.2 研究概况 ............................................................................................................ 2

§ 1.3 本文主要研究内容及研究框架 ........................................................................ 6

第二章相关理论方法简述 ............................................................................................ 9

§ 2.1 超网络简述......................................................................................................... 9

§ 2.1.1 超网络基本理论简述.................................................................................. 9

§ 2.1.2 超网络相关研究简述................................................................................ 11

§ 2.2 中心性简述 ...................................................................................................... 18

§ 2.2.1 程度中心性简述........................................................................................ 19

§ 2.2.2 中间中心性简述........................................................................................ 20

§ 2.2.3 接近中心性简述........................................................................................ 23

§ 2.2.4 特征向量中心性简述................................................................................ 26

§ 2.3 本章小结 .......................................................................................................... 29

第三章突发事件信息传播超网络（EIDSN）的建模与分析 .................................... 30

§ 3.1 突发事件信息传播超网络的建模 .................................................................. 30

§ 3.1.1 基于信息渠道的突发事件信息传播超网络建模.................................... 30

§ 3.1.2 基于信息流的突发事件信息传播超网络建模........................................ 32

§ 3.2 突发事件信息传播超网络模型分析 .............................................................. 35

§ 3.2.1 突发事件信息传播超网络中子网的差异性分析.................................... 35

§ 3.2.2 突发事件信息传播超网络的特性与应用分析........................................ 37

§ 3.2.3 突发事件信息传播超网络邻接矩阵及网络直观图的构建.................... 38

文档加载中……请稍候！
如果长时间未打开，您也可以点击刷新试试。

下载文档到电脑，查找使用更方便

15 积分 4人已下载

立即下载 VIP免费下载

摘要：

摘要近年来突发事件发生较多，随着信息技术的快速发展和社会系统的日益复杂，与突发事件相关的信息传播也越来越快速、隐蔽，如果我们不能及时了解相关舆情或谣言在网络中的传播规律，并对其进行有效的引导与管理，任其扩散爆发，将可能造成灾难性的后果。从现有研究看，较多工作偏重于同属性节点网络中的信息传播研究，然而，对突发事件信息传播系统而言，同质信息传播网络模型具有很大的片面性和局限性。作为一个复杂的社会系统，突发事件信息传播系统中涵括了多种不同类别的要素，在同质或异质要素之间还存在着复杂的相互关联。如此复杂的结构形态和构成，很难用单一要素模式的网络模型进行描述，更难以进行全面深入的分析与探究。另外，从突发...

展开>> 收起<<

基于中心性的突发事件信息传播超网络中重要节点的评判研究.PDF

共114页,预览10页

还剩页未读，继续阅读

基于中心性的突发事件信息传播超网络中重要节点的评判研究

相关推荐

开通VIP享超值会员特权

作者详情

相关内容

推荐作者

热门标签

举报选择: