一种基于进化BP神经网络的文本分类算法

提出并实现了一种结合BP神经网络和遗传算法的文本分类算法,根据遗传算法能够快速优化网络权重以及摆脱BP算法局部极点困扰的能力,提出一种改进的遗传算法确定网络拓扑结构和训练网络的方法.最后对设计的分类器进行了开放性测试,实验结果表明该分类器显著地提高了文本分类的查全率和查准率.

A Text Classification Algorithm Based on Evolutionary-BP Neural Network

This paper designed a text classification algorithm combined with BP-neural network and genetic algorithm.According to the genetic algorithm's ability of optimizing network weights and out of the trouble of local extreme points,an improved genetic algorithm was put forward to determine the network topology construction and the method of training network.Finally,an open test was done on this classifier.The experiment results show that the classifier can remarkably improve the recall and precision of text cla...