改进词频分类器集成的文本分类算法

互联网容纳了海量的文本信息,文本分类系统能够在给定的类别下,自动将文本分门别类,更好地帮助人们挖掘有用信息.介绍了基于词频分类器集成文本分类算法.该算法计算代价小,分类召回率高,但准确率较低,分析了导致准确率低的原因,在此基础上提出了基于改进词频分类器集成的文本分类算法,改进后的算法在文本权重更新方面做了参数调整,使得算法的准确率有显著提高,最后用实验验证了改进后算法的性能.实验结果表明,基于改进词频分类器集成的文本分类算法不仅提高了分类的准确性,而且表现出较好的稳定性.

A text classification algorithm that uses an improved term frequency classifier ensemble

The internet contains massive amounts of textual information.Text classification systems are essential to sort this chaotic mass into desired categories,enabling people to find the type of information they seek.The authors analyzed a text classification method that is based on a term frequency classifier ensemble.The method has low computational costs and high recall rate.Unfortunately the method lacks precision.The causes of its poor precision were analyzed.An improved method was then proposed.In the impro...