介绍了基于Ｋｏｈｏｎｅｎ和ＢＰ神经网络结合的Ｉｎｔｅｒｎｅｔ网上文本学习算法。它采用向量空间模型对文本进行编码，利用 Ｋｏｈｏｎｅｎ网络的自组织特性和ＢＰ网络的非线性特性进行学习。经过训练，算法能够有效地对输入文本进行判断，给出一个评价等级，标识出文本和用户兴趣的相关程度，从而为基于Ｉｎｔｅｒｎｅｔ的信息过滤、智能浏览等处理提供基础。 This paper introduces an internet online text learning algorithm based on Kohonen and BP neural network. It uses a vector space model to encode the text, using the self-organizing features of Kohonen network and the non-linear features of BP network. After training, the algorithm can effectively judge the input text, give a rating, and identify the correlation between the text and the user’s interest, so as to provide the basis for Internet-based information filtering and intelligent browsing.