The Design And Realize Of Real-time Network Opinion Monitor System

Nowadays, more and more people pay attention to network opinion monitor system.With the development of internet, mostly social news and opinions are spreaded by network. In recent years, there are many publish opinions, for example, the melamine, the swill-cooked dirty oil, the plasticizer and so on. The internet has made great contributions to the spread of these public opinions. Network opinion monitor system is going to find these related opinions in time, and this will help government and company to deal with the opinions better, when opinions are in prime stage.On the other hand, paper news are becoming network data, so it is possible to get and analyse opinion data.In this paper, I realized the network opinion monitor system. This system can get and analyse the news、forum、blog and micro-blog data, and pick up the opinion intelligently. I divide opinion system into 4 parts. The first part is information crawler, which will fetch news data in internet. The second part is information extraction. It will extract the title, publish time and main content from the web page. However, different web pages have different format, so it is hard to extract information. I designed a algorithm, which is available to most web pages. The third part is relativity analysis. In this part, I will filter a lot of web pages, and only keep the news which is user concerned. The forth part is opinion display. I will display the opinions with picture, chart, table and classification, which can help improve the experience of using opinion monitor system.In this paper, I make a lot of effort to improve the performance of opinion system, such as real time response, more accurate result, scalability and so on. Now my system has been completed totally, and it has been passed by the real project test. So the opinion monitor system is going to be on work in no time.