Design And Implementation The Public Opinion Monitoring System POMS

With the rapid development of the Internet,the public opinion information related to various enterprises have exploded,and the public opinion information have played an important role in the business development of enterprises.Therefore,there are lots of research value in public opinion information.After years of development,the enterprise public opinion monitoring system has made some progress in helping enterprises acquire,search,analyze and monitor the public opinion information related to enterprises,and has been widely used.However,in the face of more and more kinds of self-media data such as websites,forums and microblogs,how to obtain relevant information effectively,in the process of extracting the body of data preprocessing,taking both the speed and the quality into account,Information;how to personally recommend to enterprise users the information they are interested in is still an unsolved problem.Based on the above two issues,the main contents of this paper include:1.Constructed the framework of the public opinion monitoring system,design and implement data acquisition module,event analysis module,searching system module,article recommendation module and user management module of the public opinion monitoring system.Aiming at the problem that it is hard to balance the speed and quality of the content extraction method in data preprocessing,this paper studies the text extraction algorithm,analyzes the advantages and disadvantages of the algorithm based on the label path ratio and the text extraction algorithm based on the line block distribution detect the location of the body of the page and body position storage,and then extract the body of the method.In view of the problem that the search results are not personalized enough in the full-text search module of public opinion monitoring system,a more accurate search result is obtained by using the content-based recommendation method and the user model with decreasing user interest over time.This paper constructs a public opinion monitoring system for enterprises through incremental parallel data crawling,good user modeling and content classification.The systems and modules have been applied in practical laboratory systems and have certain innovativeness.Good application prospects.