类比推理与神经网络

Analogy Reasoaing and Neural Network

目前国内外对于类比推理的研究多侧重于逻辑推理，而较少涉及基于数字特征的推理。本文试图在类比推理中引入神经网络的概念，通过趋势预测模型选择的实例，阐述ＮＮ方法在知识表示、类比检索、类比映射和类比转换中的应用。

篗ost studies about anafogy reasoning are now concentrating on fogic reasoning，but fewer on numerical feature based reasoning.In this article，neural network theory is in-troduced into analogy reasoning，and through the examples of trend forecast model choice，the a pplication of neural network theory is expressed in the konwledge representation,analogy retrievation,analogy mapping and analogy adaptation.