基于知识的模型自动选择策略  

Knowledge-based Automatic Model Selection Strategy

作　　者：戴超凡[1] 冯旸赫[1] 

机构地区：[1]国防科技大学信息系统与管理学院,长沙410073

出　　处：《计算机工程》2010年第11期170-172,共3页Computer Engineering

基　　金：国家部委基金资助项目

摘　　要：模型自动选择是决策支持系统智能化发展的必然要求。针对目前实用算法较少的现状,提出一种模型自动选择策略。基于知识框架描述模型,根据事实库和知识库提取相应规则生成推理树,结合经验和专业知识实现模型自动选择。实验结果表明,该策略具有较高的命中率。It is necessary for the intelligent Decision Support System（DSS） to automatic select models, but there is less the available strategy. The knowledge-based automatic model selection strategy is proposed in this paper. The model is represented by knowledge frame and the reasoning trees are established according with the fact base and knowledge base, and then the model selection scheme is provided by integrating the experience and expert knowledge. Experimental result shows that the strategy has high hit rate.

关 键 词：模型自动选择 决策支持系统 框架 贝叶斯推理 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]