首页> 外文期刊>Journal of intelligent & fuzzy systems: Applications in Engineering and Technology >Approximate reasoning mechanism: internal, external, and hybrid
【24h】

Approximate reasoning mechanism: internal, external, and hybrid

机译:近似推理机制:内部,外部和混合

获取原文
获取原文并翻译 | 示例
获取外文期刊封面目录资料

摘要

We suggest an input-output system point of view of the mechanism in generalized modus ponens (GMP). The antecedent and consequent in the premise are treated as an input-output pair. The other premise and the conclusion are treated as another pair. It is suggested that the two pairs are related to each other in the sense that they are generated by the same system. With the input-output system aspect, we propose to categorize the previous approximate reasoning mechanism to be either internal or external. More insights of the internal and external mechanism are explored. In addition, a new hybrid mechanism is introduced for the OMP. We suggest a way to justify that one pair is reasoned based purely on the other. This is through the comparison of the truth values of the two pairs. The hybrid mechanism preempts that the truth value of one input-output pair, which also can be viewed as a fuzzy inference, should be identical to the other. This is coined as the Principle of Equal Truth Value (PETV). Some GMP performance, which people would often expect in a realistic GMP but is never rigorously guaranteed in other mechanism, is a natural consequence of the hybrid mechanism. Another interesting result of the hybrid mechanism is that it not only indicates what can be achieved, but also what can not be done in the reasoning. This appears to be more aligned with human experience: If the basis of reasoning (in the GMP, it is Premise 2) is too far away from the occurrence (i.e., Premise 1), then there is nothing one can say about the consequence (i.e., Conclusion) due to the occurrence.
机译:我们建议从输入输出系统的角度来看广义模态(GMP)机制。前提和前提因此被视为输入-输出对。另一个前提和结论被视为另一对。建议将这两个对彼此关联,因为它们是由同一系统生成的。在输入输出系统方面,我们建议将先前的近似推理机制归类为内部或外部。探索内部和外部机制的更多见解。另外,为OMP引入了新的混合机制。我们建议一种证明一对完全基于另一对推理的方法。这是通过比较两对值的真值而得出的。混合机制优先考虑一个输入输出对的真值(也可以视为模糊推论)应该与另一个相同。这被称为平等真理价值原则(PETV)。人们通常会在现实的GMP中期望某些GMP性能,但是在其他机制中从来没有严格地保证这些性能,这是混合机制的自然结果。混合机制的另一个有趣的结果是,它不仅指示可以实现的目标,而且还指示推理中不能完成的目标。这似乎更符合人类的经验:如果推理的基础(在GMP中,是前提2)与事件的发生距离太远(即前提1),那么就没有人可以说出后果了(即结论)由于发生。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号