首页 | 本学科首页   官方微博 | 高级检索  
     

基于神经网络与贝叶斯的混合文本分类研究
引用本文:陈世立,高野军. 基于神经网络与贝叶斯的混合文本分类研究[J]. 电脑开发与应用, 2006, 19(12): 27-29,32
作者姓名:陈世立  高野军
作者单位:中国运载火箭技术研究院研究发展中心,北京,100076;北京航天长征飞行器研究所,北京,100076
摘    要:采用向量空间模型(V SM)描述文本,利用隐性语义索引(LSI)技术进行特征重构与降维,构造了BP神经网络文本分类器。将贝叶斯分类技术与前者结合构造了一种混合文本分类器。实验结果表明混合分类器分类准确度和分类速度得到提高。

关 键 词:文本分类  BP神经网络  贝叶斯  隐性语义索引
文章编号:1003-5850(2006)12-0027-04
收稿时间:2006-07-13
修稿时间:2006-07-132006-10-08

Research on Hybrid Text Classifier based on BP Neural Network and Bayesian Approach
Chen Shili. Research on Hybrid Text Classifier based on BP Neural Network and Bayesian Approach[J]. Computer Development & Applications, 2006, 19(12): 27-29,32
Authors:Chen Shili
Abstract:Documents are represented by vector space model (VSM).Latent semantic indexing(LSI) is used to construct new features and reduce dimensionality.A text classifier based on BP neutral network is designed.Bayesian method is applied to former classifier and a hybrid text classifier is built.The experimental results show that the hybrid classifier has higher classification precision and is more efficient than other classifiers.
Keywords:text classification   BP neural network  Bayes   LSI
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号