Chinese text classification based on VC-dimension and BP neural network

Abstract

Chinese text classification based on BP neural network is an important research topic in information retrieval. In this paper, by analyzing the method of the classification based on BP, we propose that used VC-dimension to certain the number of neurons in hidden layers, describe and Analysis of the algorithm. Then, evaluating results by precision rate, recall rate and F1. Experimental results show that the proposed method has high accuracy and efficiency. © 2012 Springer-Verlag.