Research overview of microblog analysis

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论



    Microblog is one of the important social information communication platform. Because of its characteristics of easy operation and fast spread, people can directly and quickly express their attitude to emergencies, public figures, hot products and daily life through microblog. In order to utilize the vast microblog information, it needs to combine various microblog analysis methods to discover the potential value of information. This paper first reviews the current research of microblog analysis field and puts forward the independent research and development system of microblog analysis, then explores the future research direction of microblog analysis.First of all, this paper introduces two dominating technical methods of microblog analysis, including microblog open platform and web crawler technology. People can have convenient and prompt access to microblog information resources via open interface of microblog, such as microblog content, user reviews, user personal details, the number of fans, the number of attention, etc.. But it also has many limitations. For example, it can only be grabbed for limited times every hour and it is not open for all the resources information. Utilizing the web crawler technology, more information resources can be obtained. For example, the information acquisition technology based on general web crawler can cover a wider range and based on the focused crawler, the predefined theme can be crawled selectively.Secondly, this paper presents the hot issues of current microblog, including analysis on users' behavior and microblog content. Research on microblog user behavior includes: 1) Research on information communication network. Its purpose is to predict future propagation, which uses Gephi visualization tools to present propagation path under microblog diffusion, propagation range and the key of forwarding nodes, etc.. 2) Research on propagation factors. It reveals the possible causes of information dissemination by analyzing users' behavior. 3) Research on user's influence. Different scholars give different methods of measurement, while accurate evaluation of user's influence should take into account many factors, such as the number of fans, the forwarding number, mentioned numbers, reply, and social relationship, etc.. Analysis of the microblog contents includes: 1) Microblog texts pretreatment, including words segmentation and stop words. 2) Microblog hot topic detection. Its methods include statistical method of word frequency and text clustering methods, which are beneficial to the improved detection effect of hot topics without considering the characteristics of dynamic topic evolution. 3) Sentiment analysis, also known as opinion mining, has always been a hot issue in the research field of microblog. It can use microblog expression image to extract emotional words and combine with the method of constructing the semantic dictionary and machine learning to classify the sentiments of microblog, finally judge sentiment polarity of microblog. It can be used for public opinion monitoring, business forecasting and product selection, etc..Thirdly, we put forward We-Reading, an independent research and development system of microblog analysis, and introduce three main modules including sentiment analysis, geographical distribution and propagation graph expansion. Sentiment analysis module classifies users' comments sentiment based on the method of sentiment lexicon. Regional distribution module extracts the geographical position information of users and does statistical analysis in order to show the spread distribution of microblog nationwide. Propagation graph module uses visual means to represent the microblog broadcast of information diffusion, such as forwarding relations, forwarding layer, and forwarding scope, etc..Finally, this paper summarizes the challenging problems that microblog faces from two aspects of technology and applications. Technical analysis of microblog can break interface resource constraints, and improve the efficiency and precision of microblog analysis system. Practical analysis of microblog expands event monitoring, management application and commercial application.


刘 滨,张静远,刘 强,赵静阳,李 寒,徐巍巍.微博分析研究综述[J].河北科技大学学报,2015,36(1):100-110

  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
  • 收稿日期:2014-10-26
  • 最后修改日期:2014-11-25
  • 录用日期:
  • 在线发布日期: 2015-01-22
  • 出版日期: