语义检索模型中的词元扩展算法研究
|
摘要
【目的/意义】传统的信息检索技术主要是基于关键词匹配的信息推送,该方法容易出现漏检和误检的情况。语义检索通过语义分析获得用户真正的检索意图,实现精准检索。【方法/过程】本文在对语义检索的原理和模型进行描述的基础上,提出了基于本体概念树模型的词元扩展算法,通过对词元的语义相似性、语义相关性进行计算,得出词元的语义关联度,关联度超过一定阈值的词元的集合即为扩展后的词元集。【结果/结论】该方法既考虑了具有继承关系的词元间的语义相似性,也考虑了具有相同属性词元间的语义关联度,结论更具参考价值。
【Purpose/significance】The traditional information retrieval technology is mainly based on keyword matching. In this method, the results are prone to missed detection and false detection. Semantic retrieval, through semantic analysis,can obtain the user's true search intent and achieve accurate retrieval.【Method/process】In this article, the authors describe the principle and model of semantic retrieval, propose a lexical expansion algorithm based on ontology concept tree model. By calculating the semantic similarity and semantic relevance of the lexical, obtain the semantic relevance of the lexical. The set of tokens whose relevance exceeds a certain threshold is the expanded token set.【Result/conclusion】This method considers the semantic similarity between lexical elements with inheritance relations, and also considers the semantic relevance between lexical elements with the same attribute. The conclusion is more reasonable.
【Purpose/significance】The traditional information retrieval technology is mainly based on keyword matching. In this method, the results are prone to missed detection and false detection. Semantic retrieval, through semantic analysis,can obtain the user's true search intent and achieve accurate retrieval.【Method/process】In this article, the authors describe the principle and model of semantic retrieval, propose a lexical expansion algorithm based on ontology concept tree model. By calculating the semantic similarity and semantic relevance of the lexical, obtain the semantic relevance of the lexical. The set of tokens whose relevance exceeds a certain threshold is the expanded token set.【Result/conclusion】This method considers the semantic similarity between lexical elements with inheritance relations, and also considers the semantic relevance between lexical elements with the same attribute. The conclusion is more reasonable.
引文
1 ABDELKRIM BOURAMOUL,MOHAMED-KH-IREDDINE KHOLLADI,BICH-LIEN DOAN.How ontology can be used to improve semantic information retrieval:the image finder tool[J].International Journal of Computer Applications,2011,21(9):48-54.
2 Agent:An Ontology-based WWW broker to select ontologies[EB/OL].http://oa.upm.es,1998-08-23.
3阳广元.基于本体的语义检索研究现状与热点分析[J].图书馆工作与研究,2017,(10):24-30.
4徐丽.基于语义的信息检索技术研究[D].石家庄:石家庄铁道大学,2014.
5董文清.基于框架网络本体的语义检索模型[D].太原:山西大学,2010.
6刘建飞.领域本体的构建及语义检索模型研究[D].长春:吉林大学,2013.
7唐守利.基于本体的云服务语义检索模型研究[D].长春:吉林大学,2016.
2 Agent:An Ontology-based WWW broker to select ontologies[EB/OL].http://oa.upm.es,1998-08-23.
3阳广元.基于本体的语义检索研究现状与热点分析[J].图书馆工作与研究,2017,(10):24-30.
4徐丽.基于语义的信息检索技术研究[D].石家庄:石家庄铁道大学,2014.
5董文清.基于框架网络本体的语义检索模型[D].太原:山西大学,2010.
6刘建飞.领域本体的构建及语义检索模型研究[D].长春:吉林大学,2013.
7唐守利.基于本体的云服务语义检索模型研究[D].长春:吉林大学,2016.