Department of Industrial Engineering and Management, Tokyo Institute of Technology, Tokyo 152-8552, Japan.
IEEE Trans Cybern. 2013 Jun;43(3):1059-72. doi: 10.1109/TSMCB.2012.2223671.
The two most important issues of expert systems are the acquisition of domain experts' professional knowledge and the representation and reasoning of the knowledge rules that have been identified. First, during expert knowledge acquisition processes, the domain expert panel often demonstrates different experience and knowledge from one another and produces different types of knowledge information such as complete and incomplete, precise and imprecise, and known and unknown because of its cross-functional and multidisciplinary nature. Second, as a promising tool for knowledge representation and reasoning, fuzzy Petri nets (FPNs) still suffer a couple of deficiencies. The parameters in current FPN models could not accurately represent the increasingly complex knowledge-based systems, and the rules in most existing knowledge inference frameworks could not be dynamically adjustable according to propositions' variation as human cognition and thinking. In this paper, we present a knowledge acquisition and representation approach using the fuzzy evidential reasoning approach and dynamic adaptive FPNs to solve the problems mentioned above. As is illustrated by the numerical example, the proposed approach can well capture experts' diversity experience, enhance the knowledge representation power, and reason the rule-based knowledge more intelligently.
专家系统最重要的两个问题是获取领域专家的专业知识以及表示和推理已确定的知识规则。首先，在专家知识获取过程中，由于跨职能和多学科的性质，专家小组常常表现出不同的经验和知识，产生不同类型的知识信息，如完整和不完整、精确和不精确、已知和未知。其次，模糊 Petri 网 (FPN) 作为一种很有前途的知识表示和推理工具，仍然存在一些不足之处。当前 FPN 模型中的参数无法准确地表示日益复杂的基于知识的系统，并且大多数现有知识推理框架中的规则无法根据命题的变化（如人类认知和思维）进行动态调整。在本文中，我们提出了一种使用模糊证据推理方法和动态自适应 FPN 进行知识获取和表示的方法，以解决上述问题。通过数值示例可以看出，所提出的方法可以很好地捕捉专家的多样性经验，增强知识表示能力，并更智能地推理基于规则的知识。
