一种基于向量空间模型的Web本体自动分类方法

A vector space model based automatic Web ontology classification method

在分析了Web本体的结构特征和语言学特征基础上,引入虚拟文档的概念,定义整个本体的虚拟文档为与主题相关的vocabularies的虚拟文档的组合.以虚拟文档中的词条作为Web本体分类的特征项.基于RDF图不容忽视的图状特性,在构造自RDF图本体的词汇依赖图(vocabulary dependency graph)之上采用相关基于图的排序算法,得到与构造本体虚拟文档相关的vocabularies对于该本体的重要性权值,进而计算特征项的权值.

This paper propose a vector space model based automatic Web ontology classification method,which takes Web ontology′s structure and linguistic features into account at the same time.It treats the words of the virtual document of the whole Web ontology as the features for classification.As a collection of weighted words,the virtual document of the whole Web ontology is constructed by combining all the virtual documents of the vocabularies that occur in the RDF graph and are not belonged to the built-ins provided by ontology language.The way of term weighting is based on vocabulary dependency graph by applying graph-based ranking algorithm on it to get the importance score of the related vocabularies firstly and then to calculate the weight of each term.VDG is constructed from RDF graph which model the dependencies among vocabularies within an ontology.