基于社交网络的网络舆情研究(任务书,开题报告,论文19000字)
摘要
近年来，随着互联网的快速发展以及移动手持设备的普及，互联网成为了人们了解和获取事件发展信息、发表意见和评论的主要媒介。现在，现实生活中一旦发生社会事件，就会经由网络得到迅速且广泛的传播，得到网民广泛的讨论和转载。针对同一个事件，每一个网民都有不同的看法，这导致他们之间很容易产生矛盾或争执。并且随着事件的不断发酵和扩散，网民群体极化现象的发生几率变高，只有明确网络舆情的发展态势，对其加以正确的引导，才可以避免舆情危机的发生，进而维护社会的稳定，维持社会正常的发展秩序。
本文以微博大数据为研究对象，主要进行了以下工作：
舆情事件概要抽取：利用TextRank算法对博文文本内容进行处理和分析，抽取出事件的摘要。
舆情事件影响力的度量：在分析事件影响力因素的基础上，依据模糊层次综合评价法综合各因素，对事件的影响力进行计算。
舆情事件分析结果可视化：基于Bootstrap搭建前端框架，利用Echarts组件分别对事件概要、传播趋势、舆论聚合进行可视化展示。
关键词：摘要抽取；TextRank；影响力；舆情分析
Abstract
In recent years, with the rapid development of the Internet and the popularity of mobile handheld devices, the Internet has become the main medium for people to understand and obtain event development information, opinions and comments. Now, once a social incident occurs in real life, it will be spread quickly and widely through the Internet, and it will be widely discussed and reprinted by netizens. For the same incident, every netizen has a different view, which leads to conflicts or disputes between them. And with the continuous fermentation and proliferation of events, the probability of polarization of netizens becomes higher. Only by clarifying the development trend of network public opinion and correctly guiding them can we avoid the occurrence of public opinion crisis and maintain social stability. Maintain a normal development order in society.
This article takes Weibo big data as the research object, and mainly carries out the following work:
(1) Summary of public opinion events: The TextRank algorithm is used to process and analyze the text content of the blog post, and extract the summary of the event.
(2) Measurement of the influence of public opinion events: Based on the analysis of the influence factors of the events, the fuzzy level comprehensive evaluation method is used to synthesize various factors to calculate the influence of the events.
(3) Visualization of public opinion analysis results: Based on Bootstrap's front-end framework, Echarts components are used to visualize event summary, propagation trends, and public opinion aggregation.
KeyWords：Extractive summarization; TextRank; Influence; Public opinion analysis
目录
摘要 Ⅰ
Abstract Ⅱ
第1章绪论 1
1.1 研究目的和意义 1
1.2 研究现状 2
1.3 研究内容与主要工作 2
1.4 本文组织结构 2
第2章舆情分析系统研究 4
2.1 需求分析 4
2.2 功能模块设计 5
2.3 系统流程 5
2.4 相关技术介绍 6
2.5 本章小结 7
第3章数据采集与预处理 8
3.1 数据采集 8
3.2 数据预处理 9
3.2.1 数据清洗 9
3.2.2 数据排序 10
3.2.3 数据分类统计 10
3.3 数据存储 10
3.4 本章小结 11
第4章数据分析与建模 12
4.1 事件摘要抽取 12
4.1.1 基于TextRank的摘要抽取 12
4.1.2 事件摘要的抽取 14
4.2 影响力指数计算 18
4.2.1 影响力指标的确定 18
4.2.2 基于模糊层次综合评价法的影响力计算 18
4.3词云图的设计与实现 22
4.3.1 功能流程 22
4.3.2 功能实现 23
4.4 本章小结 24
第5章数据可视化 25
5.1 事件概要展示模块 25
5.2 传播趋势模块 26
5.3 舆论聚合模块 29
5.4 本章小结 31
第6章总结与展望 32
6.1 总结 32
6.2 展望 32
参考文献 34
致谢 35 |
