网络舆情监测追踪系统的研究与开发
发布时间：2018-07-22 13:09
【摘要】：网络舆情形成迅速，对社会影响巨大。随着因特网在全球范围内的飞速发展，网络媒体已被公认为是继报纸、广播、电视之后的“第四媒体”，网络成为反映社会民生的主要载体之一。已日益成为舆情产生和传播的重要领域，网络舆情在社会生活中扮演着越来越重要的角色。其直接性，突出性，偏差性使网络舆情双刃剑角色的作用越发明显。为了加强网络管理和监控，，如何更好的开展网络舆情信息的监测、追踪与分析，已经成为目前政府、企业、学校等机构面临的首要现实问题。网络舆情监测追踪系统可以实现针对网络中海量舆情信息自动实时的采集与分析，有效地解决以传统的人工方式对舆情监测、追踪和分析的难题。 传统的基于文本聚类的网络舆情热点追踪算法，在处理海量网页时，文本聚类速度过低，聚合结果较差。本文提出了一种基于关键词提取的网络舆情热点追踪方案，并根据微博、新闻、论坛和博客的不同特点分别设计了热点分析模型。通过在笔者开发的啄木鸟网络舆情监测追踪系统上的实验表明，该方案行之有效，热点分析模型识别热点准确率高。 网络舆情监测与传统搜索引擎最大的区别就在于其时效性强，滞后的信息毫无价值。及时性包括两个方面，采集性的及时性与信息分析的高效性。本文提出了首先在海量的网站中根据影响力的排名，选取有限的主流权威站点作为首要信息采集源，将有限的资源充分利用，最终提升监测效率。 最后，根据网络的发展现状，本文给出了一个具体的应用系统啄木鸟舆情监测追踪系统。引入近几年来针对网络舆情监控方面进行系统建设的过程中出现的问题以及解决具体问题的思路，全面分析了设计建设的目标和原则，同时深入阐述了各项功能的详细作用及最终实现，为及时准确预警网络突发事件和全面应对分析做出一些自己的贡献。
[Abstract]:The network public opinion forms rapidly, the influence to the society is enormous. With the rapid development of the Internet in the world, the network media has been recognized as the "fourth media" after newspapers, radio and television. The network has become one of the main carriers reflecting the people's livelihood. Network public opinion plays a more and more important role in social life. Its directness, prominence, deviation make the role of network public opinion double-edged sword more and more obvious. In order to strengthen the network management and monitoring, how to better carry out the monitoring, tracking and analysis of network public opinion information has become the most important practical problem that the government, enterprises, schools and other institutions are facing. The network public opinion monitoring and tracking system can realize the automatic real-time collection and analysis of mass public opinion information in the network, and effectively solve the difficult problem of monitoring, tracking and analyzing public opinion in the traditional manual way. The traditional hot spot tracking algorithm of network public opinion based on text clustering, when dealing with massive web pages, the speed of text clustering is too low, and the aggregation result is poor. This paper proposes a hot spot tracking scheme based on keyword extraction, and designs a hot spot analysis model according to the different characteristics of Weibo, news, forum and blog. The experiment on the woodpecker network public opinion monitoring and tracking system developed by the author shows that the scheme is effective and the hot spot analysis model has high accuracy rate of hot spot recognition. The biggest difference between online public opinion monitoring and traditional search engine lies in its timeliness, lag information is of no value. Timeliness includes two aspects: the timeliness of collection and the high efficiency of information analysis. In this paper, firstly, according to the rank of influence, we select the limited mainstream authoritative sites as the primary information collection source, make full use of the limited resources, and finally improve the efficiency of monitoring. Finally, according to the development of the network, this paper presents a specific application system, woodpecker public opinion monitoring and tracking system. This paper introduces the problems in the process of system construction in the area of network public opinion monitoring in recent years and the train of thought to solve the specific problems, and analyzes the objectives and principles of the design and construction in an all-round way. At the same time, the detailed function of each function and its final realization are expounded in detail, and some contributions are made for timely and accurate warning of network emergencies and comprehensive response analysis.
【学位授予单位】：河北大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TP391.1;G206
本文编号：2137571
[Abstract]:The network public opinion forms rapidly, the influence to the society is enormous. With the rapid development of the Internet in the world, the network media has been recognized as the "fourth media" after newspapers, radio and television. The network has become one of the main carriers reflecting the people's livelihood. Network public opinion plays a more and more important role in social life. Its directness, prominence, deviation make the role of network public opinion double-edged sword more and more obvious. In order to strengthen the network management and monitoring, how to better carry out the monitoring, tracking and analysis of network public opinion information has become the most important practical problem that the government, enterprises, schools and other institutions are facing. The network public opinion monitoring and tracking system can realize the automatic real-time collection and analysis of mass public opinion information in the network, and effectively solve the difficult problem of monitoring, tracking and analyzing public opinion in the traditional manual way. The traditional hot spot tracking algorithm of network public opinion based on text clustering, when dealing with massive web pages, the speed of text clustering is too low, and the aggregation result is poor. This paper proposes a hot spot tracking scheme based on keyword extraction, and designs a hot spot analysis model according to the different characteristics of Weibo, news, forum and blog. The experiment on the woodpecker network public opinion monitoring and tracking system developed by the author shows that the scheme is effective and the hot spot analysis model has high accuracy rate of hot spot recognition. The biggest difference between online public opinion monitoring and traditional search engine lies in its timeliness, lag information is of no value. Timeliness includes two aspects: the timeliness of collection and the high efficiency of information analysis. In this paper, firstly, according to the rank of influence, we select the limited mainstream authoritative sites as the primary information collection source, make full use of the limited resources, and finally improve the efficiency of monitoring. Finally, according to the development of the network, this paper presents a specific application system, woodpecker public opinion monitoring and tracking system. This paper introduces the problems in the process of system construction in the area of network public opinion monitoring in recent years and the train of thought to solve the specific problems, and analyzes the objectives and principles of the design and construction in an all-round way. At the same time, the detailed function of each function and its final realization are expounded in detail, and some contributions are made for timely and accurate warning of network emergencies and comprehensive response analysis.
【学位授予单位】：河北大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TP391.1;G206
【参考文献】
相关期刊论文 前10条
1 廉捷;刘云;;网络舆情中的信息预处理与自动摘要算法[J];北京交通大学学报;2010年05期
2 罗坤瑾;;网络舆情监控与构建公共信息空间[J];当代传播;2012年06期
3 施聪莺;徐朝军;杨晓江;;TFIDF算法研究综述[J];计算机应用;2009年S1期
4 刘毅;;略论网络舆情的概念、特点、表达与传播[J];理论界;2007年01期
5 姜胜洪;;网络舆情热点的形成与发展、现状及舆论引导[J];理论月刊;2008年04期
6 曾润喜;;网络舆情信息资源共享研究[J];情报杂志;2009年08期
7 钱爱兵;;基于主题的网络舆情分析模型及其实现[J];现代图书情报技术;2008年04期
8 殷卫东;朱晓华;赵俊凯;;网络舆情技术浅析[J];信息网络安全;2012年03期
9 张寿华;刘振鹏;;网络舆情热点话题聚类方法研究[J];小型微型计算机系统;2013年03期
10 范学贵;;网络社情民意 推进社会主义民主政治建设探析[J];协商论坛;2012年02期
本文编号：2137571
本文链接：https://www.wllwen.com/kejilunwen/sousuoyinqinglunwen/2137571.html
