Research of Model Intelligent Selection Based on Rule Reasoning

Decision-making model management is the key to success and practicality of decision-making support system. Model selection plays an important role in model management as foundation. Expert system theory is adopted to model selection. Model selection knowledge is obtained for each node of model classification tree, represented by architecture and production rules. Model selection knowledge tree is setup to make knowledge structuring and systematic. On the basic of model selection knowledge tree, build up model selection reasoning machine adopting framework reasoning and local-preference searching strategy. The method is applied successfully to product quality defects warning model of a management platform for a steel company.