基于知识发现的范例推理系统

1 引言范例推理(Case-Based Reasoning,CBR)是近十几年来人工智能中发展起来的区别于基于规则推理的一种推理模式,它是指借用旧的事例或经验来解决问题、评价解决方案、解释异常情况或理解新情况。CBR兴起的主要原因是传统的基于规则的系统存在诸多的缺点,如:在知识获取问题上存在困难,对于处理过的问题没有记忆而导致推理效率低下,不能有效地处理例外情况,整体性能较为脆弱等等,而CBR恰好能解决以上问题。

Case-Based Reasoning System Based on Knowledge Discovery

Nowadays the research and exploitation of the case-based system are getting more and more attention. Case-Based Reasoning (CBR) is a strategy for solving the object cases based on the source cases that are prompted by the object ones. CBR is not only a psychological theory for human knowledge, but will be a new cornerstone of the intelligent computer system technology. The case-based system is adopted in more and more application fields in order to obtain better results, especially in the fields with ill-defined and no expert knowledge. But there is a lot of knowledge required in CBR, and we are also faced with the same knowledge acquisition bottleneck as in the expert systems-Data Mining (DM) and Knowledge Discovery in Database (KDD) are just the most useful means to solve this kind of problem in order to make the knowledge acquisition more automated . In this paper, we discuss the data mining technology in CBR, especially we raise knowledge discovery in case base (KDC) and discuss this concept in detail. Finally , the structure of CBR based on DM is put forward.