基于贝叶斯粗糙集和混合专家模型的CBR 系统

建立一个完整的案例推理系统,提出一种高效的案例检索方法和一种案例调整策略.在案例检索过程中,提出一种基于贝叶斯粗糙集的属性权重确定算法,在此基础上利用最邻近法检索出与当前案例最相似的一组案例作为参考.使用检索出的相似案例训练分层混合专家模型,并用微粒群算法优化模型参数,实现了对案例的调整.采用实际转炉生产数据进行仿真,结果表明了该案例推理系统的有效性.

Bayesian rough set and mixture experts model based CBR system

An integrated case-based reasoning system is built, and an efficient case retrieval method and a case adjust method
are proposed in the system. In the case retrieval process, an algorithm to compute attributes weight is presented based on
improved Bayesian rough set model. Then, the most nearest neighbor method is used to retrieve a group of similar cases as
the reference of current furnace. In case adjust process, a hierarchical mixture of experts model is trained with this group of
similar cases, then particle swarm optimization algorithm is adopted to optimize the parameters. A simulation experiment
is implement with practical data from a steel plant. The results show the effectiveness of the proposed case-based reasoning
system.