专题研究

社交网络舆情中意见领袖主题图谱构建及关系路径研究——基于网络谣言话题的分析

展开
  • (1吉林大学管理学院 长春 130022;2吉林大学大数据管理研究中心 长春 130022; 3吉林大学国家发展与安全研究院 长春 130022)
王晰巍,女,1975年生,吉林大学管理学院教授,博士生导师,吉林大学大数据管理研究中心主任、吉林大学国家发展与安全研究院主任。 张柳,女,1992年生,吉林大学管理学院博士研究生。 韦雅楠,女,1980年生,吉林大学管理学院博士研究生。 王铎,男,1990年生,吉林大学管理学院博士研究生。

网络出版日期: 2020-03-20

基金资助

本文系国家社科基金重大项目“大数据驱动的社交网络舆情主题图谱构建及调控策略研究”(项目编号:18ZDA310)的研究成果。

Research on Thematic Map Construction and Relationship Path of Opinion Leaders in Social Network Public Opinion: An Analysis Based on Internet Rumor Topics

Expand
  • (1School of Management, Jilin University, Changchun, 130022; 2Research Center for Big Data Management, Jilin University, Changchun, 130022; 3National Institute of Development and Security, Jilin University, Changchun, 130022)

Online published: 2020-03-20

摘要

[目的/意义]大数据驱动的社交网络舆情传播中网络谣言关系路径主题图谱可视化分析,可以为相关舆情管理部门有效监管和正确引导社交网络舆情起到重要的指导和推动作用。[方法/过程]文章采用网络爬虫对社交网络舆情传播中网络谣言话题进行数据获取,构建网络谣言基于意见领袖主题图谱的实体及用户关系,确定意见领袖的关系路径分析参数;以新浪微博“重庆大巴坠江·非女司机逆行导致”话题为例,使用开源知识图谱工具Neo4j对数据进行主题图谱构建,利用Cypher语言对意见领袖的传播效率、传播路径和关键节点影响力进行分析。[结果/结论]数据结果表明,在微博网络谣言传播中,意见领袖知识图谱是一种有向网络关系图,意见领袖节点具有极强的影响力,普通用户节点的传播极易受到意见领袖节点的影响,意见领袖是舆情谣言话题传播中的关键人物;同时社交网络舆情的知识图谱整体呈现二级传播态势,单个意见领袖节点不具备左右谣言话题传播的能力,意见领袖节点与普通用户节点的壁垒随着网络谣言舆情的不断传播将在一定程度地慢慢弱化,使社交网络舆情向更为扁平及无序的方式传播。

本文引用格式

王晰巍, 张柳, 韦雅楠, 王铎 . 社交网络舆情中意见领袖主题图谱构建及关系路径研究——基于网络谣言话题的分析[J]. 情报资料工作, 2020 , 41(2) : 47 -55 . DOI: 10.12154/j.qbzlgz.2020.02.006

Abstract

[Purpose/significance] The visual analysis of the topic map of network rumor relation path in the big data- driven public opinion propagation of social network can play an important guiding and promoting role for the relevant public opinion management departments to effectively supervise and correctly guide the public opinion of social net? work.[Method/process] In this paper used the web crawler to obtain data on Internet rumor topics in the dissemination of public opinions in social networks, and based on the topic map of opinion leaders to construct the entity and user rela? tionships of Internet rumors, determined the relationship path analysis parameters of opinion leaders. Taking Sina Wei? bo topic of "chongqing bus's fall into the river·non-female driver's retrofitting" as an example. Neo4j, an open source knowledge graph tool, is used to construct the theme graph of data, and Cypher language is used to analyze the commu? nication efficiency, communication path and influence of key nodes of opinion leaders. [Result/conclusion] Data re? sults show that knowledge map of opinion leaders is a kind of directed network relation graph in Weibo network rumor propagation. Opinion leader nodes have a strong influence, and the communication of ordinary user nodes is extremely susceptible to the influence of opinion leader nodes. Opinion leader is the key figure in the propagation of rumors. At the same time, the knowledge map of social network public opinion presents a secondary communication trend. A single opinion leader node does not have the ability to control the spread of rumor topics. The barrier between opinion leader nodes and ordinary user nodes will slowly weaken to a certain extent with the continuous spread of network rumors and public opinions, so that social network public opinions will spread in a flatter and disorderly way.

文章导航

/