| 一种向量空间模型中对特征项的改进方法及应用 |
| |
| 引用本文： | 潘国清.一种向量空间模型中对特征项的改进方法及应用[J].湖南工程学院学报(自然科学版),2009,19(2). |
| |
| 作者姓名： | 潘国清 |
| |
| 作者单位： | 无锡市广播电视大学信息与艺术系,无锡,214021 |
| |
| 摘 要： | 针对传统VSM方法中选取特征项的不足之处,在文本相似度计算研究领域中中,提出了一种新的特征项选取方法--用语言片段作为特征项.介绍了选取语言片段作特征项方法的实现过程,分析了该方法的必要性与可行性,通过计算在受限语境下文本的相似度的实验,测得系统判分正确率为85.24%.与同类系统进行对比实验,发现该方法在受限语境下表现稍好.
|
| 关 键 词： | 语言片段 特征项 向量空间模型 语义 相似度 |
An Improved Method of Feature in VSM and Its Application |
| |
| PAN Guo-qing.An Improved Method of Feature in VSM and Its Application[J].Journal of Hunan Institute of Engineering(Natural Science Edition),2009,19(2). |
| |
| Authors: | PAN Guo-qing |
| |
| Institution: | PAN Guo-qing (Dept.of Information & Art,Wuxi Radio & TV University,Wuxi 214021,China) |
| |
| Abstract: | In View of the inadequacies of traditional VSM methods of selecting characteristics,this article proposes a new method of selecting the characteristics by using the fragment of language.The article describes implementation process of selecting the fragment of language and analyzes the necessity and feasibility of the method.The experiment of calculating the similarity of limited context texts shows that the correctness of the system reaches 85.24%.Compared with similar systems,the method is better in limite... |
| |
| Keywords: | language-fragment characteristics VSM semantic similarity |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
