基于WordNet概念向量空间模型的文本分类  

WordNet-based Concept Vector Space Model for Text Classification

作　　者：张剑[1] 李春平[1] 

机构地区：[1]清华大学软件学院,北京100084

出　　处：《计算机工程与应用》2006年第4期174-178,共5页Computer Engineering and Applications

摘　　要：文章提出了一种文本特征提取方法,以WordNet语言本体库为基础,以同义词集合概念代替词条,同时考虑同义词集合间的上下位关系,建立文本的概念向量空间模型作为文本特征向量,使得在训练过程中能够提取出代表类别的高层次信息。实验结果表明,当训练文本集合很小时,方法能够较大地提高文本的分类准确率。In this paper,we design and implement an automatic text classification system,aiming at improving the accuracy of text classification.ln current existing automatic text classification systems,the content of text is described by N-dimension feature vector model,but the approaches for establishing the model have great influence on the accuracy of text classification.Vector Space Model（VSM）,as one of the most effective approaches,describes a document as orthogonal term vectors.The assumption of the VSM approach is that the semantic relation between terms is ignored.But in the real world,semantic relations between terms usually exlst,such as synonymy and hypemymy-hyponymy,etc.Here we introduce a novel approach,based on WordNet,for describing a text by establishing concept vector space model.In our approach,we can extract the high-level information on categories during training process by replacing terms with synonymy sets in WordNet and considering hypemymy-hyponymy relation between synonymy sets.We carry on a series of experiments to compare our approach with the term-based VSM approach.The results show that our approach could improve the accuracy of text classification especially when the size of trainning set is small.

关 键 词：文本自动分类 WORDNET 概念向量 向量空间模型 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]