Difference Analysis of Research Topics in a Specific Domain Based on Different Content Levels

Abstract

[Purpose/Significance] This paper aims to explore whether there are differences in the title and abstract, citation content, and full-text content on research topic, and analyze whether the topic content in the title and abstract can reveal the research content of the full text and the effect of the citation content on the content of the citing literature, so as to provide theoretical support for analyzing the research content of the full text based on the title and abstract of the literature. [Method/Process] This paper conducts an empirical study using Chinese journal papers in the field of COVID-19, extracts feature words from the titles and abstracts, citations and full-text contents of the literature, uses the clustering algorithm to cluster the feature words, and then uses manual interpretation to identify the research topics, and conducts a comparative study to analyze the topic differences among the three parts. [Results/Conclusions] The results show that: the research topics are different in the title and abstract, citation content and full-text content of the literature; compared with the title and abstract, the full text contains more topic content, but the difference in the topic content is small, so the topic content in the title and abstract can be used to represent the research content of the full text; the content of the citation is related to the topic of the citing literature, and they can complement each other.

Keywords

• covid-19|feature word extraction|word clustering|topic analysis|topic model