The Relationship between the Scale of Fine-grained Knowledge Entities in Academic Articles and Articles' Academic Impact

Li, Haochuan; Zhao, Yi; Zhang, Heng; Yang, Yukai; Zhang, Chengzhi

Permalink

https://hdl.handle.net/2142/122808 Copy

Description

Title

The Relationship between the Scale of Fine-grained Knowledge Entities in Academic Articles and Articles' Academic Impact

Author(s)

• Li, Haochuan
• Zhao, Yi
• Zhang, Heng
• Yang, Yukai
• Zhang, Chengzhi

Issue Date

2024-03-20

Keyword(s)

• Full-text of academic article
• Fine-grained knowledge entities
• Entity extraction
• Academic impact of article

Abstract

In recent years, research interest has increased in exploring various factors affecting the academic impact of articles. Earlier studies primarily examined the relationships between external aspects of articles, like authors, institutions, and funding, and their academic impact. Fine-grained knowledge entities within academic articles can reflect themes and methods of research, suggesting that leveraging these elements might influence the scholarly impact of articles. In this study, focusing on the field of natural language processing (NLP), we categorized four types of knowledge entities from academic articles: method entities, dataset entities, tool entities, and metric entities. We counted the number of unique knowledge entities in each category for each academic article, defined as the scale of the knowledge entities. Subsequently, we explored how the scale of the knowledge entities across these categories correlates with the academic impact of the articles. This study demonstrates a significant and positive correlation between the scale of the knowledge entities in four categories and academic impact. Notably, the scale of method entities shows the strongest link with academic impact, followed by dataset entities. These insights can guide researchers in understanding how referencing specific knowledge entities can boost the academic impact of their academic articles.

Publisher

iSchools

Series/Report Name or Number

iConference 2024 Proceedings

Type of Resource

Other

Genre of Resource

Conference Poster

Language

eng

Handle URL

https://hdl.handle.net/2142/122808

Copyright and License Information

Copyright 2024 is held by Haochuan Li, Yi Zhao, Heng Zhang, Yukai Yang, and Chengzhi Zhang. Copyright permissions, when appropriate, must be obtained directly from the authors.

Owning Collections