Abstract:Aiming at the problem of text classification, one text classification method based on the probabilistic neural network ( PNN ) and learning vector quantization ( LVQ ) is proposed. The text features and feature values are extracted by use of TFIDF method, and text categorization feature vector are formed. In addition, classification model based on probabilistic neural network can be constructed and the learning of competitive layer network is completed by using LVQ algorithms, so the corresponding pattern vector to move closer to each other, away from the other modes, thereby realizing text classification. The experimental results show that the method in the text classification performance with very good results, and not only has good classification accuracy, but also shows a good learning efficiency.