广义规则表示及其推理算法

THE REPRESENTATION OF BROAD SENSE RULES AND ITS REASONING ALGORITHM

一般的知识规则表示法,既没有考虑前提条件对结论的不平等性,又没有考虑时间对结论成立的影响,而这些问题对某些领域是非常重要的.另一方面,在不精确推理中,以往的系统都推到终结点之后,才根据阈值判定此结论是否成立,因而影响了推理速度,为此,本文在实践的基础上,提出了“重要度”、“时间因子”及“分层阈值”等概念和它们的算法及推理算法.

The general representation of knowledge rules considers neither the une-qual influence of premises on the conclusion nor the influence of time on the conclusion, which are both very important for some fields. On the other hand, in inexact reasoning, the con-clusion is determined to be true or false according to the threshold after the reasoning. This influences the reasoning speed. Based on practice, this paper presents some new concepts, such as importance degree, time factor, and hierarchy threshold, their algorithms, and their rea-soning algorithms.