Applicayion And Reseach On Case-Based Reasoning In Enterprise Evaluation Based On Rough Set

With the internationalization of China's Economy, financial organizations are facing more and more problems on challenge and their efficiency need to be improved. Based on advanced computer technology and powerful network technology, we can set up decision support system in financial organizations in order to gain decision information quickly, to analyze the quality of credit, and to consummate the control of risk. This is a helpful method to meet challenges and improve competition.The national 863 project is a decision system of enterprise credit risk evaluation based on middleware. Reasoning machine of artificial intelligenc expert system and secondary level reasoning mechanism case-based reasoning and rule-based reasoning are used in this system. This system meets the requirements of modernization. This dissertation focuses on two algorithms. One is used to compute weights, which is more objective, and the other is used to compute similarity which more suits for reality. Both of them are based on reasoning machine and rough set. Acorrding to these contents above, it did serverl works as follow.First, it puts forward an algorithm for computing weights based rough set. This solution of identifying weight values is more objective, and it avoids subjective and blindness of previous resolution.Second, it puts forward an algorithm of computing similarity based on previous work when some of characteristic items can not be identify. And it works well on this situation.Finally, it presents some problems exsiting in original system, and gives resolution of how to perfect it. Moveover, it has provided the direction and the precious experience for the further research.