基于本体和关联数据的引文分析方法研究

Research on the Citation Analysis Method Based on Ontology and Linked Data

将本体、关联数据技术与引文分析结合,提出了基于本体和关联数据的引文分析方法。以期利用语义技术来揭示学术文献之间的引用关系,实现语义出版环境下全新模式的引文分析。首先,选取CSSCI数据库引文分析领域的高被引文献组成实验数据集,并对引文题录数据进行抽取和预处理。接着,构建题录引文本体BCO对引文数据进行规范化描述,并结合语义仓储工具发布为引文关联数据。最后,构建不同的SPARQL检索式对引文信息进行多维度抽取,对数据集引文年代、引文作者和引文期刊的分布情况进行具体分析,并进行跨维度分析,实验验证基于本体和关联数据的引文分析方法的可行性。研究发现基于本体和关联数据的引文分析方法具有一定的可行性和实用性,值得科研人员参考借鉴。

This paper proposed a new citation analysis method based on ontology and linked data by way of combining ontology, linked data technology and citation analysis. The purpose was to make use of semantic technology to reveal the citation relationships among academic documents, and to create a new citation analysis method in the semantic publishing environment. First of all, the highly cited literature in the citation analysis field of CSSCI database was selected to compose the experiment data set. Then, a Bibliographic Citation Ontology(BCO) was constructed to normalize the citation data and to publish RDF triples as citation linked data. Finally, the authors constructed different SPARQL queries to extract the bibliographic citation information in multiple dimensions. They also analyzed the citation times, citation authors, citation journals, and multi-dimensional situation in detail, hoping to verify the feasibility of this method. The citation analysis method based on ontology and linked data has certain feasibility and practicability. It is worthy of consideration by scientific researchers.