View in EDS

Interdisciplinarity in academic research: a keyword-based approach to knowledge diversity.

Saved in:
Bibliographic Details
Title: Interdisciplinarity in academic research: a keyword-based approach to knowledge diversity.
Authors: Sun, Jiajia1 (AUTHOR) sunjiajiacn@qq.com, Li, Yajing2 (AUTHOR) jingjingcn@qq.com
Source: Scientometrics. May2026, Vol. 131 Issue 5, p3679-3698. 20p.
Subjects: Interdisciplinary research, Interdisciplinary approach to knowledge, University research, Content analysis, Heterogeneity
Abstract: Interdisciplinary measurement quantifies the degree of knowledge integration within academic research and provides a basis for assessing knowledge diversity. Existing methodologies predominantly rely on citations and author collaborations, approaches that overlook textual content and require high-quality data. This paper presents a text-based method for measuring the interdisciplinary nature of research papers through keyword diversity. The approach imposes lower data-quality requirements while capturing knowledge diversity from a content perspective. First, keyword diversity datasets are constructed. Second, these data are used to measure paper interdisciplinarity. Metadata for papers published between 1900 and 2018 are collected from the Web of Science database. Two experiments are conducted: one based on reference lists and the other based on keywords. The controlled experiments evaluate integrated indices, including the Rao–Stirling index, True Diversity, and DIV*. The results demonstrate that the keyword-based method generally achieves higher recall than the reference-based method, which can substantially increase measurement error and operational complexity. In addition, the Rao–Stirling and True Diversity indices approximately follow a normal distribution, whereas DIV* exhibits monotonicity. A correlation analysis on large-scale statistical data distributions is also conducted, revealing significant correlations between the keyword-based and reference-based measures across the three indices (correlation coefficients of 0.97, 0.90, and 0.74, respectively). These results confirm the feasibility and effectiveness of the keyword-based approach. [ABSTRACT FROM AUTHOR]
Copyright of Scientometrics is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Engineering Source
Description
Abstract:Interdisciplinary measurement quantifies the degree of knowledge integration within academic research and provides a basis for assessing knowledge diversity. Existing methodologies predominantly rely on citations and author collaborations, approaches that overlook textual content and require high-quality data. This paper presents a text-based method for measuring the interdisciplinary nature of research papers through keyword diversity. The approach imposes lower data-quality requirements while capturing knowledge diversity from a content perspective. First, keyword diversity datasets are constructed. Second, these data are used to measure paper interdisciplinarity. Metadata for papers published between 1900 and 2018 are collected from the Web of Science database. Two experiments are conducted: one based on reference lists and the other based on keywords. The controlled experiments evaluate integrated indices, including the Rao–Stirling index, True Diversity, and DIV*. The results demonstrate that the keyword-based method generally achieves higher recall than the reference-based method, which can substantially increase measurement error and operational complexity. In addition, the Rao–Stirling and True Diversity indices approximately follow a normal distribution, whereas DIV* exhibits monotonicity. A correlation analysis on large-scale statistical data distributions is also conducted, revealing significant correlations between the keyword-based and reference-based measures across the three indices (correlation coefficients of 0.97, 0.90, and 0.74, respectively). These results confirm the feasibility and effectiveness of the keyword-based approach. [ABSTRACT FROM AUTHOR]
ISSN:01389130
DOI:10.1007/s11192-026-05651-9