《计算机应用研究》|Application Research of Computers

基于语义关系约束和词语关系信息的句向量研究

Sentence vector based on semantic relationship constraints and word relationship information

免费全文下载 (已被下载 次)  
获取PDF全文
作者 夏小强,邵堃
机构 合肥工业大学 计算机与信息学院,合肥 230009
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2019)07-024-2023-04
DOI 10.19734/j.issn.1001-3695.2018.01.0029
摘要 针对现有的句向量学习方法不能很好地学习关系知识信息、表示复杂的语义关系,提出了基于PV-DM模型和关系信息模型的关系信息句向量模型(RISV)。该模型是将PV-DM模型作为句向量训练基本模型;然后为其添加关系信息知识约束条件,使改进后的模型能够学习到文本中词语之间的关系,并将关系约束模型(RCM)作为预训练模型,使其进一步整合语义关系约束信息;最后在文档分类和短文本语义相似度两个任务中验证了RISV模型的有效性。实验结果表明,采用RISV模型学习的句向量能够更好地表示文本。
关键词 句向量; RISV模型; PV-DM模型; 关系信息; 预训练
基金项目
本文URL http://www.arocmag.com/article/01-2019-07-024.html
英文标题 Sentence vector based on semantic relationship constraints and word relationship information
作者英文名 Xia Xiaoqiang, Shao Kun
机构英文名 School of Computer Science & Information,Hefei University of Technology,Hefei 230009,China
英文摘要 In view of the fact that the existing sentence vector learning method can not well learn the relational knowledge information and express the complicated semantic relation, this paper proposed a relational information sentence vector model(RISV) based on the PV-DM model and the relational information model. This model used the PV-DM model as the basic model of sentence vector training, and then added the knowledge constraint of relational information to make the improved model could learn the relationship between the words in the text and used the RCM model as pre-training model to further integrate the information of the semantic relationship constraints, and finally validated the validity of the RISV model in two tasks: document classification and short text semantic similarity. The experimental results show that sentence vectors learned by RISV model can better represent the text.
英文关键词 sentence vector; RISV model; PV-DM model; relationship information; pre-training
参考文献 查看稿件参考文献
 
收稿日期 2018/1/17
修回日期 2018/3/9
页码 2023-2026
中图分类号 TP183
文献标志码 A