文本层次分类系统的研究.pdfVIP

文本层次分类系统的研究 高波赵政 (天津大学仿息学院计算机系，天津 3ω072) E-mail: tju_gb@ 摘 要 文章提出了层次分类模型，将类别按相似程度形成…棵树形结构，对文章分类时泛…层一层逐层比较的，这样 就使得文本分类时文本与类别之间的比较次数大为减少，同时由于大的类别的特征之间的区别比较明泉，因此又能在一 定程度上提高文本分类的精准率。考虑到一篇文章的标题和正义对决定义章所处的类别上所起的作用是不闷的，文中将 标题和正义分开处理。还有在进行特征边排时将 mDF 和 MI 结合起来，这也是该文的创新之处。实革命结果表明，层次分 类的方法在速度上比一般分类快 15%左右，而精准且在又有一定程度的提高。 关键诩文本分类向受空间精准牟层次分类 文就编号 lω2-8331-(2006) 11-0176-03 文献标识码 A 中阁分类号 TP311 Research 00 Text Hierarchical Classificatioo System Gao Bo Zhao Zheng (lnstitute of Electronic and Information Engineering ,Tianjin University ,Tianjin 300072) Abstract: We bring forward the level …classìfied model ,which puts together the alike class to become the construction of a tree form b租sed on their similarity ,80 when deciding the class of a text ,the comparison is from layer to layer ,and this makes the times of comparison decreωing greatly ,at the same time ,because of the greater distinction of big category ,again on a certain degree increasing the precision of classification.In consideration of the function of the headline of article is different to the text in deciding its clas日， we treat them separately in computing the value of similarity ,and still when calculating the eigenvalue we use both TFIDF and MI algorithm ,these are a11 the innovation of this thesis.The result of the research indicates that ,the speed of level-classified algorithm is 15 percent quicker than the general algorithm ,again on a certain degree can ìncrease the precisìon of classification. Keywords: text classification ,vector space ,precision ,hierarchical classification 1 文本分提简介 向量距离分类法、贝叶斯方法、K 最近邻方法、支持向量机算法 1.1 文本分类概述 和神经网络算法等等，简略介绍一下最常用的三种算法: