论文部分内容阅读
在分析了当前文本分类中常用方法的基础上 ,提出了一种新的分类模型 .该模型是对人的分类过程的一种模拟 .在已有的英语语义词典及大量训练集的基础上 ,应用机器学习、数据挖掘等技术进行知识获取并最终形成若干个概念推理网 .对待分类的文档可以激活相应的网络 ,同时传播推理以决定其类别的归属 ,试验表明 :该方法具有较高的分类正确率与召回率 .
On the basis of analyzing the commonly used methods in the current text classification, a new classification model is proposed, which is a simulation of the human classification process.Based on the existing English semantic dictionary and a large number of training sets, Applying techniques such as machine learning and data mining to acquire knowledge and finally forming a number of conceptual reasoning networks, the documents to be classified can activate the corresponding networks and disseminate reasoning to determine the belonging of the categories. The experiment shows that the method has higher classification Correct rate and recall rate.