Web上支持产品设计的灵活的实例抽取技术

CBD设计支持系统采用 Web上基于实例推理与基于规则推理相结合的方法。由于知识获取瓶颈的存在 ,文中提出 ,修改过程应当由人借助于机器推理的结果来完成 ,而不是直接由机器来完成。提出的灵活的实例模板技术 ,解决了系统实例结构的可进化 ;提出的用户视图技术 ,较好地解决了实例的客户化问题。并讨论了有助于实例抽取的实例形式化问题

Flexible Case Retrieval on the Web for Product Design Support

Case based design (CBD) is the application of case based reasonning (CBR) to the design process. There are two problems that hindered the application of CBD. One problem lies in the modification process. Due to the fact that knowledge acqusitition is still the bottleneck of the design process, this paper proposes that modification should be done by designers, rather than by machines directly and thoroughly. And the machine inference results can be used as suggestion to designers in the modification. Another problem relates to case evolution and customization. The paper proposes the agile case template technique and the user's template technique that overcome the problem perfectly. Case formalization that is helpful to case indexing is also discussed.