中文摘要：

对于中文文本分类问题，使用了一种新的RBF神经网络算法．这一方法通过高斯径向基函数，使用k均值推导出隐藏项的中心点及宽度，并将由隐藏层得到的输出结果合并起来，从而得到分类结果．试验证明，这种算法的准确率、召回率、F测量的值都很高，得到的分类效果很好．

英文摘要：

As to the problem of Chinese text classification, a new method of RBF network is used. Through Gaussian radial basis function, k-means is employed to get the centers and widths of hidden unit. Then outputs of hidden layer are combined. Finally, classification results are obtained. The experiment proves that the precision, recall and F-measure of this method are very high, and classification effect is perfect.