Research On Discovery Of Forest Products Business Web Information Source

To solve the "information overload" problem and meet the demand of specific topic information, the technology of topic-oriented web information integration becomes an important research topic. The web information integration technology can get information from different sources together to provide vertical information service. The trade information in forest products business website is an important forestry information resource. The integration of trade information is the basis of forest products business information service. Previous research has preliminarily integrated some information from different websites. But the information sources are found manually, which spend lots of time. An automatic method of discovering forest products business information source is need.In this work, the current technologies of website discovery were introduced. Then a new method of forest product business information source discovery was presented. The method divided the problem into web search and website classification. Web search aimed to start from "seed websites", download a few pages and look for external links. Website classification was use to tell forest products business websites apart. When doing website classification, a feature keyword database was established and an improved VSM based on key resources was presented. A website classifier based on SVM was introduced. Finally the forest products business information source module was designed and achieved. About110forest products business websites were discovered in the experiment. The experiment performed well, which means the effect of the method was confirmed.