A Study And Application On Knowledge-Based Decision Support Model

《A Study and Application On Knowledge-based Decision Support Model》 is a branch of "Management Decision-Oriented Intelligent Systems Study of Isomeric Knowledge representation and knowledge Management" which is sponsored by National Natural Science Foundation of China.A New DSS model system must support model integration , model choice and knowledge access, in addition ,the emphasis should be put on model representation and model operation these two aspects . With the development of computer technology, AI technology was imported into model system, and theory of identifying natural language, knowledge representation and model selection methods have new development.Traditinal decision-making problem often can't be expressed well and can't be understood by computer. Some complex decision making problems which have bad structure or ill structure can not be resolved . Facing to this situation, at firstly this paper identify problem based on natural language, second participate words by the best prefix arithmetic ,third analyses syntax by rule base for getting problem 's objective ,situation and data. This system improved the ability of system'sintercommunion between human and computer and the system flexibility and the system intelligence.View model as a kind of knowledge, modeling model based on knowledge, Express model using object-oriented knowledge representation and express knowledge of knowledge base using rule for sharing knowledge, reusing, combining different models and assembling other knowledge information , it is the base of transferring ill structure to well structure .On researching model access, Model selection has main function in model management. The paper focus on model choose, it realized automatism choosing model based on problem's objective and situation by heuristic rule.At last based on research work, this paper designed and implemented a decision support model management system. The system composed of problem resolving and system management .The system implemented intercommunion between human and computer based on natural language, transferring ill structure to well structure, model choose and getting the result. Combining model base and knowledge base makes this system can resolve complex decision making problem. It is well used in assistant decision-making domain showed that the research did well in improving the ability and intelligence of resolving complex decision making problems.