Extracting Relations from Chinese Web Documents Using Kernel Methods

Extracting instances of a given target relation from a given Web page corpus seems to be the basic work to exploit nearly endless source of knowledge which provided by the World Wide Web. In this paper, we present an automated system which could extract instances of an arbitrary given binary relation from Chinese Web documents in domain of football games. Different syntactic sources are combined by Kernel methods. And dependency path is used as pattern information when discover relation pairs. Moreover, composite kernels are developed to show the usefulness of different syntactic sources. Experimental results show the effectiveness and benefits of our approach.