Research On Chinese WEB Information Acquisition System Oriented To Cooperative Partner Selecting

The paper roots in the project"Research on Cooperative Partners Selecting Based on Web Mining"(No: 70572070), which is funded by National Science Foundation. This project study the theory and method of applying web mining on internet to look for cooperative partners, offering more chances for choosing cooperative partner to core enterprise, shortening time of establishing virtual enterprise, improving the market response speed of core enterprise.Illumined by the project, the author expects to use the correlated text disposing technique, such as web pages collecting, text filtering, information extraction and so on, to gain enterprise attribute information automatically, such as corporation name, corporation scale, corporation throughput, corporation contact information and so on. This can provided a lot of information material to core enterprise, which also can be offered lots of support for cooperative partners selecting.Firstly the author presents the total frame of Chinese Web information acquisition system oriented to cooperative partner selecting, and analyzes the key technology of implementing the system, which is web pages collecting based on meta-search engine, corporation homepages filtering based on samples'common feature and corporation homepages information extraction based on schema. The three key technology were introduced in detail. According to the thinking proposed, an archetype system is developed by programming computer. The feasibility of the thinking proposed is verified and accuracy is proved by the developed system.