BBS准实时舆情监测技术研究与实现

发布时间：2018-06-10 10:42

本文选题：高校论坛 + 信息采集　；参考：《华中科技大学》2012年硕士论文

【摘要】：舆情监测是当前各级部门的一项重要工作。通过舆情监测，，获取舆论的准确动向，已经成为必须进行的工作。BBS作为当前重要信息传播渠道，在舆论传播中起了很重要的作用，研究专门针对BBS这个特定领域的舆情监测也是十分有意义的工作。目前，对于BBS舆情研究多集中在两个方面，一是研究对校园网络舆论的引导策略，二是研究BBS舆情监测系统。已有的舆情监测系统多采用基于通用搜索引擎技术的信息采集分析方法，此方法属于主动采集信息的方法，信息采集较全，但存在信息采集周期较长，重复内容较多的问题，不能满足新形势下实时、高效、准确的新需求。因此，提出了一种能够准实时监测BBS舆情信息的技术。该技术主要从数据采集、数据预处理和数据分析三个方面设计系统架构。首先在对当前武汉地区的BBS架构特征进行提取的基础上，针对BBS架构的特点提出了准实时采集数据的方案。然后采用HtmlParser工具对采集回来的数据进行预处理，同时采用Solr为数据库搭建搜索引擎。最后从热点信息提取和舆情信息提取及预警分析三个方面进行数据挖掘，得到完整的检测系统。系统能够准实时对BBS进行监测，准确获取指定内容，高效检索采集数据，进而挖掘热点事件。研究不仅能够应用于高校舆情监测，也可方便的扩展到主流论坛系统的舆情监测。目前系统已经给有关部门使用了一段时间，反映效果较好。
[Abstract]:Public opinion monitoring is an important work of departments at all levels at present. Through public opinion monitoring to obtain the accurate trend of public opinion, has become the work that must be carried out. BBS as the current important information dissemination channel, has played a very important role in the dissemination of public opinion. At present, the research on BBS public opinion is focused on two aspects, one is to study the guiding strategy of campus network public opinion, the other is to study the BBS public opinion monitoring system. The existing monitoring system of public opinion mostly adopts the information collection and analysis method based on the general search engine technology. This method belongs to the method of actively collecting information, and the information collection is more complete, but there is the problem of long period of information collection and more repeated content. Can not meet the new situation under the real-time, efficient, accurate new requirements. Therefore, a near-real-time monitoring technology for public opinion information of BBS is proposed. The system architecture is designed from three aspects: data acquisition, data preprocessing and data analysis. Firstly, based on the feature extraction of current BBS architecture in Wuhan, a quasi-real-time data acquisition scheme is proposed for the characteristics of BBS architecture. Then the HtmlParser tool is used to preprocess the collected data, and Solr is used to build the search engine for the database. Finally, data mining is carried out from three aspects: hot spot information extraction, public opinion information extraction and early warning analysis, and a complete detection system is obtained. The system can monitor the BBS in real time, obtain the specified content accurately, retrieve the collected data efficiently, and then mine the hot events. The research can not only be applied to the monitoring of public opinion in colleges and universities, but also can be conveniently extended to the monitoring of public opinion in the mainstream forum system. At present, the system has been used by relevant departments for a period of time.
【学位授予单位】：华中科技大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TP393.09;TP391.3

【参考文献】