| In a geometric series increase the social background of the rapid development of the Internet technology and Internet users, Internet users increasingly used to express wishes and opinions on the issue of national political ecology, social construction, the management of government through the network platform, especially the forums, microblogging and other network platform has become all kinds of information, news release, dissemination, sharing the mainstream media. For government departments, how to become particularly important for effective monitoring of Internet information.This paper on the research study of existing public opinion monitoring system and Linux system, web crawler, MySQL database, Jieba participle technology based, the currently popular public opinion monitoring system design concept and try to build the district public opinion monitoring analysis system are combined, for under the Web2.0 environment realize public opinion information collection, processing, analysis, reporting and other functions in the Internet public opinion monitoring and analysis platform, for the relevant departments to fully grasp the network of public opinion, strive to take the initiative to dispose of the right to provide the basis for effective analysis, lay a good decision information support for the leadership of the correct decision.For the traditional artificial search there are slow and inaccurate, this paper builds the a public opinion monitoring system based on Linux operating system, analysis system, the system is mainly with web crawling, data storage, data analysis and alert public opinion display function. System to area existing double home owners forum as the experimental object, through the actual operation of the system, users on the forum of new themes and the latest replies in the content of real-time monitoring, through the data analysis function to get the page posting information to judge the current user’s new theme and new replies existence relates to politically sensitive and residential tenants mainly concern about the hot words, and the after word frequency analysis or blacklist sink ratio of successful information in the form of mail is to push the alarm. Finally, this paper will test the system, to ensure that the system can be applied to the actual work. |