A CG-based method to construct knowledge base from the text in water conservancy domain

Aiming to provide semantic information for decision supporting system and constructing a domain knowledge base in which knowledge is represented in a formalized form so as to be understood by machine, this paper suggest a new method which is based on conceptual graph to represent knowledge formally. At first, the system architecture would have been designed, and it consists of knowledge acquirement layer, knowledge management layer and knowledge application layer. Then, the crucial technology, formalizing the domain text by conceptual graph, is expatiated by the domain ontology and knowledge is constructed synchronously. By using the specific computational tool named Prolog+CG for representing and reasoning about water conservancy texts, the validity and efficiency of this method are proven.