提出了一种基于语言神经网络的知识获取方法,该方法利用语言神经元,对具有开区域的连续输入变量,自动产生相应的语言变量输出,讨论了相应的神经网络训练和知识获取方法,所获取的知识以If-Then的规则形式表示,具有简洁、紧凑、不必进一步化简、易于理解等特点,并给出在智能教学系统中获取专家领域知识的应用实例. A method of knowledge acquisition based on language neural network is proposed. This method uses language neurons to automatically generate the output of language variables for continuous input variables with open regions. The corresponding neural network training and knowledge acquisition methods are discussed. The obtained knowledge is expressed in the form of If-Then, which is characterized by conciseness, compactness, no further simplification and easy to understand. The application example of acquiring expert knowledge in intelligent teaching system is given.