The World Wide Web is a major source of the news now. The news analysis play more and more important roles in the political decision, financial analysis, invest decision, market forecast and so on. This paper proposes a news analysis system based on the hall for workshop of metasynthetic engineering. In this system, in order to analyses online news, the news articles are active fetched, and being classified into standard categories. And then by means of adaptive clustering, the correlative news is obtained into the same cluster, helping the user to reduce the time to search the news which he cares about. Moreover, the news comments are classified into categories such as negative, positive, which are also grouped into clusters helping the experts to get the view of the common people to the news. So the news attention analysis can be obtained to help the experts find which news is the people concerned most effectively.