加权网络结构洞占据者挖掘研究及其系统实现

发布时间：2018-04-14 16:38

本文选题：加权网络 + 结构洞占据者　；参考：《安徽大学》2017年硕士论文

【摘要】：近年来,随着各种网络和计算机科学的飞速发展,网络的规模呈指数级增长,在网络中占据何种位置能够获益的想法已经得到许多人的关注。个体或者团体中的中间人可以获得丰富的信息并控制他们的网络关系,在网络中占据结构洞位置的主体可以获得丰厚的利益。作为网络结构分析的重要概念,结构洞占据者在获取网络有效信息方面起着关键的作用。结构洞理论一经提出,受到社会学、心理学、经济学等领域研究者们的重视和关注,并被广泛应用于各个领域。处在结构洞位置上的节点可以获得累加且非重复的信息,因此找到结构洞占据者并加以利用,是非常有意义的。目前,研究者们对于结构洞占据者挖掘的研究主要侧重于无权网络,对加权网络的研究较少,在实际网络中,边的权重对网络的性能有重要影响,边的权重能更真实地反映网络结构的特性。本文的主要工作包括两个方面。一,以加权网络为研究对象,将边的权重引入网络,提出一种基于加权中介中心性的结构洞占据者挖掘方法(W_BCHIS),用以挖掘加权网络中的结构洞占据者。该方法首先利用社团发现算法求出加权网络的社团,接着引入边权改进中介中心性指标,以初始化节点的重要性,通过对节点的重要性和节点占据结构洞的可能性不断地迭代直到收敛以挖掘出结构洞占据者。最后,在公用数据和实例数据上验证本文所提算法的有效性。二,本文设计并实现一个结构洞占据者挖掘系统,该系统主要包括社团划分模块和结构洞挖掘模块,其中,社团划分模块实现了社团划分算法,结构洞占据者挖掘模块实现了本文提出的W_BCHIS算法、无权的HIS算法、无权的MaxD算法和加权W_HIS等算法。同时,用户可通过系统进行不同的社团划分方法和结构洞占据者挖掘方法的实现对比分析。本文工作如下:1.介绍结构洞占据者挖掘基本理论概念和相关算法。包括结构洞的定义、度量指标及基本算法。2.针对加权网络的问题研究,提出基于加权中介中心性的结构洞占据者获取方法(W_BCHIS)。本文以加权网络为研究对象,将权重引入到中介中心性指标中,对现有的HIS算法进行改进,提出一种基于加权中介中心性的结构洞占据者挖掘算法(W_BCHIS),并将本文提出的W_BCHIS算法与其他算法在公用数据集上进行实验对比与分析,验证算法的有效性。3.设计并实现结构洞占据者挖掘系统。系统的主要功能模块包括社团划分模块、结构洞占据者挖掘模块和可视化图表呈现模块。该系统基于B/S架构,主要的开发语言采用Java,运用MVC设计模式对系统的主要功能进行实现。
[Abstract]:In recent years, with the rapid development of various networks and computer science, the scale of the network has increased exponentially.The middleman in the individual or group can obtain abundant information and control their network relationship, and the main body occupying the structure hole in the network can gain rich benefits.As an important concept of network structure analysis, structural hole occupants play a key role in obtaining effective network information.Once the theory of structural hole is put forward, it has been paid attention to by researchers in sociology, psychology, economics and so on, and has been widely used in various fields.Nodes in the position of structural holes can obtain cumulative and non-repeated information, so it is very meaningful to find the occupants of the structural holes and make use of them.At present, the researchers mainly focus on the structure hole occupant mining, but less on the weighted network. In the actual network, the weight of the edge has an important impact on the performance of the network.The weight of the edge can reflect the characteristics of the network structure more realistically.The main work of this paper includes two aspects.Firstly, taking the weighted network as the research object, introducing the weight of the edge into the network, this paper presents a method of structure hole occupant mining based on weighted intermediary centrality, which can be used to mine the occupants of the structural holes in the weighted network.In this method, the community of weighted network is obtained by using community discovery algorithm, and then the importance of initializing nodes by introducing edge weight to improve intermediary centrality index is introduced.The importance of nodes and the possibility of nodes occupying structural holes are iterated until they converge to excavate the occupants of structural holes.Finally, the effectiveness of the proposed algorithm is verified on public data and instance data.Second, this paper designs and implements a structure hole occupant mining system. The system mainly includes community partition module and structure hole mining module, among which, community partition module implements community partition algorithm.The structure hole occupant mining module implements the proposed W_BCHIS algorithm, unauthorized HIS algorithm, unauthorized MaxD algorithm and weighted W_HIS algorithm.At the same time, the user can carry on the different community division method and the structure hole occupant mining method through the system to carry on the contrast analysis.The work of this paper is as follows: 1.This paper introduces the basic theory concept and related algorithms of structure hole occupant mining.Including the definition of structural holes, metrics and basic algorithm. 2.In order to solve the problem of weighted network, this paper presents a new method of obtaining structural hole occupants based on weighted intermediary centrality.In this paper, the weighted network is taken as the research object, the weight is introduced into the intermediary central index, and the existing HIS algorithm is improved.In this paper, a new structure hole occupying algorithm based on weighted intermediary centrality is proposed. The W_BCHIS algorithm proposed in this paper is compared with other algorithms on the common data set to verify the validity of the algorithm.Design and implement the structure hole occupant mining system.The main function modules of the system include community division module, structure hole occupant mining module and visualization chart presentation module.The system is based on B / S architecture, the main development language is Java, and the main functions of the system are realized by MVC design pattern.
【学位授予单位】：安徽大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：O157.5

【参考文献】