论文部分内容阅读
一、引言当今专家系统已逐步成为人工智能中影响最大、应用最广泛的领域之一。然而,尽管专家系统技术在具有良结构的狭窄领域内表现出很高的性能,但仍存在许多问题。如:·处理困难或不常见问题时性能急剧下降;·由于系统知识库的非结构化,系统知识难以维护和修改;·系统不能积累以往问题求解的经验;·系统行为的解释仅仅是启发式推理规则的再现,缺乏理论支持,不能令人信服。之所以存在上述问题,圭要是由于缺乏
I. Introduction Today’s expert systems have gradually become one of the most influential and widely used fields of artificial intelligence. However, although expert system technology shows high performance in narrow areas with good structure, there are still many problems. Such as: • A sharp decline in performance when dealing with difficult or infrequent problems • A system’s inability to maintain and modify system knowledge due to unstructured systems repositories • A system’s inability to accumulate past experience in solving problems • An explanation of system behavior is only a matter of heuristics Reproduction of reasoning rules, lack of theoretical support, can not be convincing. The reason for the existence of the above problems is due to lack of knowledge